001-SLLR-SLLR-1992-2-SIRIMAVO-BANDARANAIKE-v.-RANASINGHE-PREMADASA-AND-CHANDANANDA-DE-SIL.pdf
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
1
SIRIMAVO BANDARANAIKEv.
RANASINGHE PREMADASA ANDCHANDANANDA DE SILVA
SUPREME COURTG. P. S. DE SILVA, C.J.
P. RAMANATHAN, J.
S.B. GOONEWARDENE, J.
P. R. P. PERERA, J. ANDA. S. WIJETUNGA, J.
PRESIDENTIAL ELECTIONPETITION NO. 1 OF 198919 JUNE 1989 TO 30 JUNE 1992
Presidential Election Petition – General intimidation – Non-compliance withprovisions of the Presidential Elections Act No. 15 of 1981 – Failure to conduct afree and fair election in accordance with the provisions of the PresidentialElections Act – Presidential Elections Act No. 15 of 1981 ss. 91 (a). 91 (b) -Interpretation of s. 91 (a) – Burden of proof- ss. 101, 102 Evidence Ordinance.
The election to the office of President of Sri Lanka was held on 19 December1988. There were three candidates namely Sirimavo R. D. Bandaranaike(Petitioner) of the Sri Lanka Freedom Party (SLFP), Ranasinghe Premadasa (1strespondent) of the United National Party (UNP) and Oswin Abeygunasekera ofthe Sri Lanka Mahajana Party (SLMP). The petitioner received 2289860 or 44.95%of the votes, the 1st respondent 2569199 or 50.43% of the votes andAbeygunasekera 235719 or 4.63% of the votes. The first respondent won by aMajority of 279339 votes. Of the eligible voters 55.32% voted. The 2ndrespondent as Commissioner of Elections declared the 1st respondent elected tothe office of President of Sri Lanka.
The petitioner by petition filed on 09 January 1989 challenged the election of the1st respondent on the following grounds.
By reason of general intimidation the majority of electors were or may havebeen prevented from electing the candidate whom they preferred under section91 (a) of the Presidential Elections Act No. 15 of 1981 (hereinafter referred to asthe Act).
2
Sri Lanka Law Reports
[1992] 2 Sri L.R.
By reason of non-compliance with the provisions of the Act relating toelections, the election was not conducted in accordance with the principles laiddown in such provisions and such non-compliance affected the result of theelection under s. 91 (b) of the Act.
By reason of “other circumstances” to wit, the failure of the Commissioner ofElections (2nd respondent) and/or certain members of his staff to conduct a freeand fair election, in accordance with the provisions of the Act as set out moreparticularly in paragraph 9 read with paragraph 8 of the petition, the majority ofthe electors were or may have been prevented from electing the candidate whomthey preferred under section 91 (a) of the Act.
The petitioner called 546 witnesses, the 1st respondent 399 witnesses and the2nd respondent 32 witnesses.
The pivotal question in this case turns on the correct interpretation of section 91(a) of the Presidential Elections Act which reads as follows:
The election of a candidate to the office of President shall be declared to be voidon an election petition on any of the following grounds which may be proved tothe satisfaction of the Supreme Court namely:-
(a) that by reason of general bribery, general treating, or generalintimidation, or other misconduct, or other circumstances, whether similar tothose before enumerated or not, the majority of electors were or may havebeen prevented from electing the candidate whom they preferred”.
This Court in its preliminary order (reported at (1989) 1 Sri LR 420,261,270) heldthat mere proof of the several instances or acts of general intimidation would notsuffice to avoid an election. In addition, the petitioner has to prove that theseseveral acts or instances had the result or consequence that the majority ofelectors were or may have been prevented from electing the candidate whomthey preferred. On the basis of instances or acts of general intimidationestablished by evidence, the Court may draw a reasonable inference therefromthat the majority of electors may have been prevented from electing the candidateof their choice. In a case of general intimidation, the question that arises is – fromthe proved acts of intimidation of electors is it reasonable to suppose that theresult of the election may have been affected? This is the true meaning of thewords 'the majority of the electors may have been prevented from electing the
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
3
candidate they preferred. But it will be open to the returned candidate to showthat the gross intimidation could not possibly have affected the result of theelection. Proof of widespread violence directed towards preventing electors fromvoting was not enough. There was tine requirement of proof of an additional anddistinct ingredient of the charge that the majority of the electors may have beenprevented from electing the candidate whom they preferred.
The petitioner’s case was one of preventive intimidation and not coerciveintimidation.
In the expression “were or may have been prevented" there is a significantdifference between the words “were” and “ may have been”. The term may wasdesignedly used because mathematical proof that the majority of electors were infact prevented, in many a case is impossible of attainment. The burden to provethat the majority of electors were in fact prevented is difficult and it is almostimpossible to produce the requisite proof.
Held:
(a) The preliminary order made by the Court is binding on the Court. Nogloss or deviation from the order is permissible. Further trial proceeded on thebasis of the interpretation placed by Court on s. 91 (a) in the preliminary order.
(b) Proof of widespread violence directed towards preventing electors fromvoting is not enough. Proof is necessary also of the additional ingredient that thegeneral intimidation had the effect that the majority of voters were or may havebeen prevented from electing the candidate whom they preferred.
In so far as a charge under s. 91 (b) is concerned a Court must reach afinding as to whether the non-compliance affected the result of the election. TheCourt then must consider the question whether the petitioner would havesucceeded but for the non-compliance.
Per G. P. S. de Silva, C.J. “for that purpose evidence of party affiliationswould be relevant and admissible, notwithstanding the secrecy provisions wouldit then be reasonable to say that the secrecy provisions do not apply to sec. 91(b) and they apply to sec. 91 (a). We think not."
The evidence of group leaders regarding party affiliations is permissibleas a mode of proof that the voters were prevented from electing the candidate of
4
Sri Lanka Law Reports
[1992] 2 Sri L.R.
their choice and will not offend the secrecy provisions. The Court may address itsmind to the pattern of voter behaviour.
On a careful consideration of the totality of the evidence relating to thecharge of general intimidation, it appears that the thrust of the J.V.P. violence wasdirected against the U.N.P. Between the period 17.09.88 and 19.12.88 (16.09.88being the date on which the Working Committee of the U.N.P. chose the 1strespondent as the candidate) as many as 413 organisers, office-bearers andsupporters of the U.N.P. were killed, and 237 were attacked. The acts of violenceagainst the U.N.P. were spread throughout 80 polling divisions in 15 electoratedistricts, whereas the anti-S.LF.P. incidents occurred in 23 polling divisions in 13electoral districts. Further the incidents against the U.N.P. were spread over alonger period of time. Numerous threats, killings and attacks on local partyorganisers and office-bearers of the U.N.P. branches at the village level resultedin a serious and irreparable setback to the organization and the campaign of the1st respondent. In addition there were resignations from U.N.P. branches byoffice-bearers and even ordinary members consequent upon threats conveyed byletters. Besides, there were threats directed at office-bearers and members of the
J.S.S. and large numbers were compelled to resign. The J.S.S. actively supportedthe U.N.P. at previous elections. It is natural that all this would have had a strongadverse effect on supporters of the 1st respondent at the Presidential election.The oral and documentary evidence establishes that the weight of the J.V.P.intimidation and violence was directed at the U.N.P. and its supporters and thishas contributed in no small measure to the low voter turn-out on 19.12.88(election day.)
The burden of proof however slight it may be is on the petitioner that theacts or instances of intimidation had the requisite effect, namely, that the majorityof electors were or may have been prevented from electing the candidate whomthey preferred. The petitioner has not succeeded in establishing that the result ofthe election may have been affected.
Accordingly the charge of ‘General intimidation’ relied on by thepetitioner as a ground of avoidance of the election fails.
Per Goonewardene J…“in terms of section 91 (a) of the PresidentialElections Act. an order to avoid an election it must be shown that it was not a freeand fair one. It is proved not to be a free and fair election, when it is proved thatthe majority of the electors were or may have been prevented from electing the
sc
Sirimavo Bandaranalke v. Ranasinghe Premadasa and Another
S
candidate whom they preferred…It is therefore, in rny view, vital to the success ofthe petitioner's case as based upon section 91 (a) of the Presidential ElectionsAct to prove as the primary requisite, that the majority of electors were or mayhave been prevented from electing the candidate whom they preferred*.
Cases referred to:
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another [1989] 1 SriLR, 240,248, 249, 250,258,259,261,263,264, 268, 269, 270,281.282.
Illangaratne v. G. £ de Silva 49 NLR169.
Abeywardene v. Ariya Bulegoda [1985] Sri LR 86.
Jayasinghe v. Jayakody [1985] 2 Sri LR 77,89.
Ratnam v. Dingiri Banda 45 NLR 145.
Pelpola v. R. S. S. Gunawardene 49 NLR 207.
Tarnolis Appuhamy v. Wilmot Perera 49 NLR 361,367,368.
North Louth Case (1911) 6 O' M & H103,124.
South Meath Case40O' M &H 130,141.
Woodword v Sarsons (1875) LR 10 C.P. 733,743,744.
Wijewardene v. Senanayake 74 NLR 97.101.
Shiv Charan Singh v. Chandra Bhan Singh and Others 1988 2 S.C.C. 12.
Hackney Case (1872) 2 0' M&H 77.
Morgan V. Simpson (1974) 3 All ER 722,725,726.
Anthony v. Seger (1780) 1 Hag. Con 9.13, (1775 -1802) All ER 549,550.
Faulkner v. Eiger (1825) 4 B & C 440.
Ashby v. White (1704) 1 Bro Pari Cas. 62.
The Drogheda Case (1869) 1 O' M & H 252.
The Bradford Case (1869) 1 O' M & H 35.
The Salford Case (1869) 1 O' M & H133.
The Stafford Case (1869) 1 O' M & H 228.
The Nottingham Case (1869) 1 O' M & H 245.
The Borough of Dudley Case (1874) 2 O' M & H 115.
North Durham Case (1874) 2 O' M & H 152.
6
Sri Lanka Law Reports
[1992] 2 Sri L.R.
The Thornbury Division of the Country of Gloucester (1886) 4 O’ M & H 63.
The Lichfield Case (1869) 1 O’ M & H 25.
The Thornbury Case (1886) 4 O' M & H 65.
The Ipswich Case (1886) 4 O' M & H 70.
London Joint Stock Bank v. Simmons (1892) AC 208.
Rex v. Dolan (1907) 2 Irish Reports 286.
R. v. Chief Constable of the Merseyside Police Ex parte Calveley and others(1986) 1 All ER 257,259.
The Dudley Case (1874) 2 O' M & H115.121.
The Bolton Case (1874) 2 O' M & H138,142.
The Norfolk Case 9 Journ. 631.
The Heyw Co. 555 (n).
The Morpeth Case 1 Doug. El. C1471.
The Pontefract Case 1 Doug El. C. 377.
The Coventry Case P & Kn 338, C & R 276.
The New Ross Case 2 PR & D 188.
The Drogheda Case W & D 206.
Attorney-General v. Prince Ernest Augustin of Hanover (1957) 1 All ER 49.
Amoah Ababio v. Turkson (1954) 1 WLR 509.
Warburton v. Loveland (1831) 2 D & CL. (HL) 489.
Vashit Narain Sharma v. Dev. Chandra AIR 1954 S. C. 513,516.
Paokai Haokip v. Rishang AIR 1969 S. C. 663. 666,667.
Election petition challenging election of President
L de Silva, PC. with Ranjith Abeysuriya, P.C., R. K. W. Goonesekera, A. A. deSilva, Sidat Nandalochana, Anil Obeysekera, Percy Wickremasekera,S. L. Gunasekera, M. W. Amerasinghe, Nihal Jayamanna, Morris Rajapakse,
Yoosuf, C. Padmasekera, Suran/ith Hewamanna, Neil Rajakaruna,S. Madawalagama and Collin Senarath, Nandadeva instructed by Nimal Siripalade Silva for petitioner.
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva, C.J.)
7
K.N. Choksy, P.C. with L. C. Senevlratne, P.C., P. Nagendra P.C.; VarunaBasnayake, P.C., Kosala Wijayatilake, P.C., S. C. Crosette Thamblah, Sunil K.Rodrigo, Jehan Cassim, Naufel Abdul Rahman, Daya Pelpola, S. I. Mohideen,Raja Dep, D. H. N. Jayamaha, S. Mahenthiran, Laksman Ranasinghe, LakshmanPerera, A.L.B. Brito Muthunayagam, Ronald Perera, A. A. M. Illiyas,Miss B. Y. Devasurendra, Miss Nilmi Yapa, instructed by S. Sunderalingam for 1strespondent.
P. S. C. de Silva, P.C. Attorney-General, Tilak Marapona, P.C. Solicitor-General,S. Aziz Additional Solicitor-General, K. C. Kamalasabayson, Deputy Solicitor-General, F. N. Goonewardena, State Counsel and Dhammika Dassanayake, StateCounsel instructed by U. R. Wijetunge, State Attorney tot 2nd respondent.
Curadvvull
1st September, 1992.
P. S. DE SILVA, C.J.
On 09th January, 1989, the petitioner Sirimavo R. D. Bandaranaikefiled this petition seeking to have the election of RanasinghePremadasa, the 1 st respondent, to the office of President of SriLanka, declared null and void. The election was held on 19thDecember, 1988. There were 3 candidates, Sirimavo R. D.Bandaranaike of the Sri Lanka Freedom Party (SLFP), RanasinghePremadasa of the United National Party (UNP) and OswinAbeygunasekera of the Sri Lanka Mahajana Party (SLMP). TheCommissioner of Elections (2nd respondent) declared the results asfollows:-
Oswin Abeygunasekera2357194.63%
Sirimavo Bandaranaike228986044.95%
Ranasinghe Premadasa256919950.43%
.Valid votes5094778
Rejected votes91445
Total polled5186223
Majority279339
Total registered votes9375742
Total polled/registered votes55.32%
8
Sri Lanka Law Reports
[1992] 2 Sri LR.
By this petition, the petitioner is challenging the election of the 1strespondent on the following grounds:-
that by reason of general intimidation, the majority of electorswere or may have been prevented from electing thecandidate whom they preferred under section 91 (a) of thePresidential Elections Act No. 15 of 1981 (hereinafter referredto as the Act);
that by reason of non-compliance with the provisions of theAct relating to elections, the election was not conducted inaccordance with the principles laid down in such provisionsand that such non-compliance affected the result of theelection under s. 91 (b) of the Act;
that by reason of "other circumstances”, to wit, the failure ofthe Commissioner of Elections (the 2nd respondent) and/orcertain members of his staff to conduct a fair and freeelection, in accordance with the provisions of the Act, moreparticularly set out in paragraph 9 read with paragraph 8 ofthe petition, the majority of the electors were or may havebeen prevented from electing the candidate whom theypreferred, under section 91 (a) of the Act.
Mr. H. L. de Silva, Counsel for the petitioner, in his closing addressabandoned charge (2) above. Therefore in these proceedings weare now concerned only with charges (1) and (3) above. The trialcommenced on 19th June, 1989, and concluded on 30th June, 1992.The petitioner called 546 witnesses, the 1st respondent 399witnesses, and the 2nd respondent 32 witnesses.
The charge of “general Intimidation” as a ground or avoidance ofthe election In terms of section 91 (a) of the PresidentialElections Act No. 15 of 1981:-
Mr. H. L. de Silva for the petitioner commenced his closingaddress before us with the submission:- “ There is in fact I venture tosubmit but one pivotal question in this case and that turns on thecorrect interpretation of section 91 (a) of the Presidential ElectionsAct”. Section 91 (a) reads as follows:-
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva, CJ.)
9
“The election of a candidate to the office of President shall bedeclared to be void on an election petition on any of the followinggrounds which may be proved to the satisfaction of the SupremeCourt, namely:-
that by reason of general bribery, general treating, or generalintimidation, or other misconduct, or other circumstances,whether similar to those before enumerated or not, the majorityof electors were or may have been prevented from electing thecandidate whom they preferred."
Mr. de Silva urged that section 91 (a) does no more than reflect theEnglish Common Law and that this was the position right from the daysof the Ceylon (State Council Elections Order in Council 1931. The sectioncontained the basic and essential principles of the English Common Lawrelating to a free and fair election. This was precisely the submission^made by Mr. de Silva at the hearing before this.Court on the preliminaryobjections (hereinafter called the preliminary hearing) filed by the 1st and2nd respondents, moving for a dismissal of the petition in limine. It washis submission then (as it was in his closing address) that the expression"general intimidation” is nowhere defined in the Act and that our Courtshave hitherto looked to the English Common Law for its meaning. Thesection is intended to protect the right of the electorate to a free and fairelection. The underlying principle is that it is in the public interest that theelection should be free and fair. In the English Common Law “generalintimidation” has a well-recognised meaning which goes back to the timeof Edward III.
In the order made by this Court on the preliminary objection(hereinafter called the preliminary order) the submissions of Mr. de Silvaare set out thus:-
“(1) The English Common Law of a ‘free and fair election’ is whatis embodied in s. 91 (a). The expression ‘majority of electors wereor may have been prevented from electing the candidate whomthey preferred’ means a majority of persons entitled to vote free ofintimidation and other pressures, were prevented or may have beenprevented from electing a candidate according to theirpreferences. The expression does not impose an additional
10
Sri Lanka Law Reports
[1992] 2 Sri LR.
burden on the petitioner, if general intimidation is established, thenecessary consequence flows – that the majority were preventedfrom electing the candidate of their choice. All that the petitionerneed establish is general intimidation. Once general intimidation isestablished, free choice goes." (The emphasis is ours)
"(2) In this view of the matter, it is not necessary to identify thecandidate whom the majority of electors would or may havepreferred. Moreover, how the voters would have voted underdifferent circumstances is impossible of proof. Unlike in the case ofthe statutory offence of undue influence where there must be anidentification of the individual affected by the intimidation in thecase of general intimidation, the identification of victims is difficultand is not necessary. Furthermore, it would violate the principle ofsecrecy of the ballot which is enshrined in Article 93 of theConstitution which enacts that ‘the voting for the election of thePresident of the Republic shall be free, equal and by secret ballot'.A voter cannot be asked for whom he would have voted, if therewas no general intimidation.” (1989 1 Sri L. R. 240 at 249 and250) <1>.
The Court then posed to itself the question, “what is the EnglishCommon Law regarding the avoidance of elections?" (at page 255)and having considered no less than seven cases dealing with theEnglish Common Law, concluded as follows:-
“From the observations made in the said cases, it seems tous to be clear that at English Common Law, where it was provedthat bribery, treating or intimidation were so general and soextensive in its operation that it prevented men of ordinary nerveand courage from going to the poll, whether or not thesuccessful candidate or his agents were responsible for thecorruption or violence, the election was set aside on the groundthat it was not free", (at page 258).
Thereafter the Court considered the rival contentions of Mr. deSilva and Mr. Choksy:-
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G.P.S.de Silva, C.J.)
11
The question arises whether s. 91 (a) of the Act embodieswhat Mr. H. L. de Silva, P.C. described as the “pure andunadulterated English Common Law” prior to 1949, or asMr. Choksy submitted, that in addition to general intimidationetc. something more has to be proved by a petitioner to have anelection avoided under s. 91 (a)." (at page 259)
Having cited the cases of lllangaratne v. G. E. de Silva li),Abeywardene v. Ariya Bulegodal3> and Jayasinghe v. Jayakodyw thecourt expressed its ruling in the following terms:-
“We agree with Mr. Choksy that more proof of the severalinstances or acts of general intimidation would not suffice toavoid an election. In addition, the petitioner has to prove thatthese several acts or instances had the result or consequencethat the majority of electors were or may have been preventedfrom electing the candidate whom they preferred’.” (at page261)
This is a very important finding in the preliminary order. The Courtdid not accept Mr. de Silva’s submission that section 91 (a) reflectsno more than the English Common Law. Mr. de Silva contended in hisclosing address that the Court was in error in reaching thisconclusion. Said Mr. de Silva in his closing address "… the SupremeCourt has misunderstood what the English Common Law on thesubject was and wrongly assumed the requirement of the EnglishCommon Law was different. The language in section 91 (a) is in facta reproduction of the English Common Law prior to 1949 and doesnot introduce an additional requirement… What the earlier Bench inthe preliminary order erroneously thought to be "additional” is also asmuch a part of the English Law” (written submissions of Mr. de Silva).
The other crucial question which the Court in its preliminary orderposed to itself is, "What is the meaning of the expression 'the majorityof electors were or may have been prevented from electing thecandidate whom they preferred?” (at page 264). The Courtproceeded -to consider several decisions, viz. Ratnam v. DingiriBanda ®, Pelpola v. R. S. S. Gunawardene (6> Tarnolis Appuhamy v.
12
Sri Lanka Law Reports
[1992] 2 Sri L.R.
Wilmot Perera m. North Louth case m, South Meath case (9> andanswered the question in the following terms:-
“So, it seems to us that on the basis of instances or acts ofgeneral intimidation established by evidence, the Court maydraw a reasonable inference therefrom that the majority ofelectors may have been prevented from electing the candidateof their choice. In a case of general intimidation, the questionthat arises is – from the proved acts of intimidation of electors, isit reasonable to suppose that the result of the election may havebeen affected*? This, it seems to us, to be the true meaning ofthe words” the majority of the electors may have beenprevented from electing the candidate they preferred". But, itwill be open to the returned candidate to show that the grossintimidation could not possibly have affected the result of theelection.” (at page 270)
This is a clear, categorical and unequivocal ruling on the keywords in section 91 (a) of the Act. It is confined in terms to a “case ofgeneral intimidation". At the argument on the preliminary objectionmuch time was spent in elucidating the meaning of these words ofcritical importance in section 91 (a). Almost all the cases cited byMr. de Silva in his closing address were cited at the hearing on thepreliminary objections. A full and comprehensive argument waspresented by both Mr. de Silva and Mr. Choksy. The argument lastedfor as long as 17 days in March and May 1989.
At the preliminary hearing Mr. de Silva strenuously contended thatsection 91 (a) has two constituent elements. First, there must be proofof certain forms of conduct or events of a specific description —general bribery, general treating or general intimidation, or othermisconduct, or other circumstances, whether similar to those beforeenumerated or not. The second limb of the section — “the majority ofthe electors were or may have been prevented from electing thecandidate whom they preferred” — connotes the consequences ofthe conduct or the events specified in the first part of the section. Theconsequence is to impede or prevent the free exercise of thefranchise. The specified forms of conduct or events, on account oftheir inherent nature, impede or prevent the free exercise of the
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G.P.S.de Silva, CJ.)
13
franchise. Once it is established that the general intimidation is of awidespread and of an all pervasive character (not of a local orisolated nature) then the inevitable consequence is that the voter’sfreedom to elect a candidate of his choice is seriously affected. Theeffect is in-built in the very nature of the forms of conduct or eventsspecifically set out in the section. The Court will draw an infeience asto the effect, having regard to the nature of the forms of conduct orevents enumerated in the section. This is a matter of legal inferencefor the Court. In other words what Mr. de Silva stressed was that theeffect or the consequence is not an independent factor or anadditional element (“something more” as stated in the preliminaryorder). In short, it is not a requirement that has to be proved aliunde.These submissions, however, did not find acceptance with the Courtin the preliminary order.
Apart from the passages in the preliminary order quoted earlier,the following passage in that order also shows that the interpretationplaced on the section by Mr. de Silva did not find favour with theCourt –
“s.91(a) of the Act states that an election will be avoided if itis ‘proved to the satisfaction of the Supreme Court that byreason of general intimidation, the majority of electors were ormay have been prevented from electing the candidate whomthey preferred'. It seems to us that it is for the petitioner to provethat there was widespread violence directed towards preventingelectors from voting. But the relief which the petitioner hasasked for under s.91(a) of the Act will be granted subject to afinding by the Supreme Court that the general intimidation hadthe effect, namely, that the ‘majority were or may have beenprevented from electing the candidate whom they preferred'. Itis a conclusion which is placed in the hands of the SupremeCourt upon a review of all the evidence." (at page 268)
This passage shows that the view of the Court was that proof ofwidespread violence directed towards preventing electors fromvoting was not enough. There was the requirement of proof of anadditional and distinct ingredient of the charge. Relief which thepetitioner has asked for under section 91(a) of the Act will be granted
14
Sri Lanka Law Reports
[1992] 2 Sri L.R.
subject to a finding by the Supreme Court that the generalintimidation had the requisite effect. In the preliminary order the Courtcited with approval the opinion of Sharvananda, C.J. in Jayasinghe v.Jayakodyw — “In order to succeed in his petition the petitioner hasgot to prove a further ingredient, viz. that the majority of the electorsmay have been prevented from electing the candidate whom theypreferred …” (The emphasis is ours)
On the other hand, it was the submission of Mr. de Silva that thesecond limb of s.91(a) was not intended to impose an additionalrequirement but it indicates the effect flowing from the “generalintimidation”. At the preliminary hearing it was the contention ofMr. de Silva that the section postulates “a composite concept” andthe latter part of the section sets out “the necessary effect of theunlawful pressures" (to use Counsel’s own words). In his closingaddress Mr. de Silva urged that “by considering this last element tobe an addition the- Court has unwittingly fallen into error and thoughtthat the Sri Lankan law on this question was in some way differentfrom the English Common Law. That this is not so may be gatheredfrom the unambiguous statement of Nagalingam, J. when he statedthe law in Tarnolis Appuhamy v. Wilmot Perera m There is not even ahint there that our law is any different from the English CommonLaw — there is no such additional element.” (Written submissions ofMr. de Silva).
We have already referred to the ruling in the preliminary order(1989 1 Sri L.R. at 270) as to the meaning of the key words “themajority of electors were or may have been prevented from electingthe candidate whom they preferred”. Mr. de Silva strongly urgedbefore us that the word “result” when used “in the judicial expositionof s.91(a) means 'the effect' or 'the consequence' which the offensiveacts have on the majority of voters’ freedom to choose the candidate… Courts have for convenience, in order to describe the requisitemagnitude or the extent to which voters were prevented from votingused the expression 'that the result may have been affected’ as ashorthand expression for describing the extent or magnitude of theaffectation. But these words are not in fact to be found in the wordsenacted by Parliament in framing para (a) of s.91. That is only ajudicial paraphrase of paragraph (a) … These words prescribe the
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. R S. de Silva, C.J.)
1S
index or the measure of the effect which the law requires theprohibited acts or disabling facts must have on the minds of thevoters”. (Written submissions of Mr. de Silva)
It appears to us, however, there is not even a hint in thepreliminary order that the Court was resorting to what Mr. de Silvacalls a “judicial paraphrase" of s.91(a) when the Court ruled that “in acase of general intimidation, the question that arises is — from theproved acts of intimidation of electors, is it reasonable to supposethat the result of the election may have been affected”. It is notwithout significance that in the very next sentence the Courtproceeded to say in clear, explicit and emphatic terms, “this it seemsto. us, to be the true meaning of the words 'the majority of electorsmay have been prevented from electing the candidate theypreferred’.” Undoubtedly the ruling is in terms that are cogent,.precise and unambiguous. There is nothing whatever to suggest inthis ruling that the words “prevented from electing the candidatewhom they preferred” mean more than saying “prevented from freelyexercising the franchise or the right to vote" as contended for byMr. de Silva.
The expression “result of the election” which occurs in the aboveruling of the Court has to.be given its plain and literal meaning, forthere is no other meaning which could be reasonably attributed,having regard to the rival contentions advanced and the context inwhich the words are used. Mr de Silva strongly relied on the fact thatin the preliminary order the Court expressly held that the “result”contemplated in s.91(b) is “the success of one candidate over theother” citing Woodword v. Sarsons(,0> but that the Court made nosuch ruling in respect of its analysis of s.91(a). This, however, doesnot mean that the Court used the words “result of the election” in asense other than its natural and ordinary meaning, namely, theoutcome of the election.
Mr. de Silva drew our attention to two submissions made byMr. Choksy for the 1st respondent in support of the preliminaryobjections. These submissions are set out in the preliminary order(1989) 1 Sri L.R. 240 at 248, and read as follows:-
16
Sri Lanka Law Reports
[1992] 2 Sri LR.
“(1) The petitioner must prove that by reason of generalintimidation, a certain result or consequence followed, namely,that ‘the majority of electors were or may have been preventedfrom electing the candidate whom they preferred’. This is animportant ingredient of the ground of avoidance in s.91(a) of theAct. If so, the petitioner must identify the candidate whom themajority of electors preferred, but were or may have beenprevented from electing by reason of general intimidation.
This is a material fact which the petitioner must prove and if itis a material fact to be proved, then it must be pleaded."
“(2) In addition, the petitioner must plead and prove how themajority of electors were or may have been prevented fromelecting the candidate whom they preferred. That is, thepetitioner must plead and prove that the majority of electorswho voted for the 1st respondent were or may have beencompelled to vote for him by reason of general intimidation, orthat the balance 45% of the electors abstained from votingbecause of general intimidatipn, and if they had voted, thereasonable probabilities are that they would have voted for her.This is a material fact which the petitioner must prove, and if soit must be pleaded".
It was the contention of Mr. de Silva that both submissions wererejected by the Court in the preliminary order. We do not agree. Thisis how the Court in its preliminary order dealt with (1) above – (1989)1 Sri L.R. 240 at 263.
“s.96(c) of the Act requires that the petition 'shall contain aconcise statement of the material facts on which the petitionerrelies'. In Wijewardena v. Senanayake, 011 H. N. G. Fernando,C.J. observed that this requirement was ‘intended to secure thata respondent will know from the petition Itself what facts thepetitioner proposes to prove in order to avoid the election andwill thus have a proper opportunity to prepare for the trial… Theterm ‘material facts' has a plain meaning in the context ofrequirements relating to pleadings, namely, facts material to
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva, C.J.)
17
establish a party’s case’. The object of the requirement is clearlyto enable the opposite party to prepare his case for the trial sothat he may not be taken by surprise. When the petitionerpleaded in paragraph 6(a) of her petition that 'there wasgeneral intimidation in consequence of which the majority of thesaid electors were or may have been prevented from electingthe candidate whom they preferred’ is there sufficientinformation given in the petition to enable the 1st respondent toidentify the candidate whom the electors were or may havebeen prevented from electing? In paragraph (1), the petitionerhas stated that she was a candidate at the Presidential Electionand ‘claims to have had a right to be returned or elected at thesaid election’. The petitioner has set out in paragraph 5 thevotes cast for each candidate and that she obtained the secondlargest number of votes. Could there be any doubt in the mindof the 1st respondent as to the identity of the candidate, who,the petitioner claims, would or may have been returned,-but forthe general intimidation?”
There is no finding here that it is not necessary for the petitioner toidentify the candidate whom the majority of electors preferred butwere or may have been prevented from electing by reason of generalintimidation. All that the Court held was that on a fair reading of therelevant averments in the petition, the test formulated in Wijewardenav. Senanayake (supra) was satisfied. In other words the pleadingswere sufficient to prevent the opposite party from being taken bysurprise. There is certainly no finding by the Court, that the petitioneris relieved of the obligation to show that she was the candidate whomthe majority of electors may have preferred but for the intimidation.
As regards (2) above, the Court having held that the petitioner’scase is one of “preventive intimidation” (and not “coerciveintimidation") and having considered Rutnam’s case, Pelpola's case,Wilmot Perera’s case (supra) and the observation of Gibson J. in theNorth Louth Case181 and of O' Brien J. in South Meath Case (supra)reached the following finding:- “In Our opinion, how the majority wereor may have been prevented from electing the candidate of theirchoice need not be specially pleaded” – at page 268. (Theemphasis is ours) In other words the Court was of the view that a
18
Sri Lanka Law Reports
[1992] 2 Sri L.R.
specific or an express plea was not necessary and the plea in thepetition in regard to the effect of the general intimidation wasadequate. We cannot agree that this finding has any furtherimplication in regard to the ingredients of the charge of generalintimidation set out in s.91(a).
The preliminary order has dealt in fair detail with the two localcases relating to a charge of general intimidation, viz. Ratnam v.Dingiri Banda and Pelpola v. R. S. S. Gunawardene (supra). Thesetwo cases are important inasmuch as they indicate the approach ofthe Court to the facts in relation to the requirement of “affectation” ofthe result. In Ratnam's case the contention of the petitioner was thathis supporters who were the Indian labourers on the estates wereprevented from going to the poll by the supporters of the contestingcandidate, M. D. Banda. Upon a review of the evidence Hearne J.expressed his finding in the following terms:- “I hold that there wasgross intimidation, that it was widespread in the areas whereMr. Ratnam had good reason, to count upon heavy voting in hisfavour, and that it may well have prevented the majority of theelectors from returning the candidate whom they preferred”. Thewords underlined above show that upon the evidence the Courtfound that there were areas within the electorate where the majority ofthe electors were supporters of the petitioner, though of course theywere prevented from going to the poll. In a narrow and unduly strictsense it could be said that this finding was reached on “opinionevidence". Nevertheless the ultimate conclusion in favour of Ratnamwas reached on the basis that the intimidation was widespread “inthe areas where Mr. Ratnam had good reason to count upon heavy
voting in his favour” If Mr. de Silva is correct in his submission that
such evidence is opinion evidence and inadmissible hearsay, thenHearne J. could not have found for the petitioner.
Pelpola's case was also a similar case where the Court specificallyaddressed its mind to the pattern of "voter behaviour”. Apart from thefact that the majority of the winning candidate was only 387 votes, theunchallenged position of the petitioner was that over one quarter ofthe electorate were Indian estate labourers against whom theIntimidation was directed. There was the evidence of the Presidentof the Ceylon Indian Congress Labour Union Committee of Mosville
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. RS.de Silva. CJ.)
19
Estate that all the Indian labourers had decided to vote for thepetitioner. Windham, J. goes on to state: "It is not unreasonable tosuppose that the Indian labourers on the neighbouring estates … hadlikewise decided to vote for the petitioner… Only 514 out of 1427recorded their votes at the Uduwela polling station … Had 400 morepersons voted and cast their votes for the petitioner the latter wouldhave won the election". (The emphasis is ours) Thus it is seen thatthe decision turned not only on the size of the majority but also on (1)the direction of the intimidation (acts of intimidation against the Indianestate labourers), (2) the probability of the Indian estate labourersvoting for the petitioner. In other words, the Court considered theevidence relating to the charge of general intimidation on the basis ofhow the electors may have voted had they the opportunity of voting -a view of the evidence which Mr. de Silva strongly urged a Court isprecluded by law from taking by reason of the secrecy provisions.These provisions no doubt preclude an elector who has voted frombeing asked for whom he voted. But do they also preclude a witnesswho claims to be an organizer or an office-bearer of a party branchfrom stating that the campaign of a candidate was disrupted bycertain incidents and that the supporters of that candidate wereafraid to go to the poll? We think not. It is very relevant to note that inboth Ratnam's case and Pelpola's case evidence of the partyaffiliation of the voters in relation to the petitioner was considered andacted upon by the Court. Similar secrecy provisions in regard to theballot were in operation at that time also, but the evidence wasadmitted. What is relevant for present purposes is that evidence ofparty affiliation was allowed as a mode of proof that the voters wereprevented from electing the candidate of their choice.
Mr. de Silva vigorously contended that the evidence led on behalfof the 1st respondent of adverse “affectation" of UNP supporters byreason of the JVP terror campaign was based on pure conjecture andspeculation: that the Court is precluded from considering suchevidence as it is founded on surmise and unpredictable factors. Wefind, however, that the case for the petitioner, as presented beforeus, was not free of evidence of this kind. A striking example was theevidence of the principal witness for the Anuradhapura electoraldistrict, K. B. Ratnayake, M.P. He said he was the authorised agent ofthe petitioner for Anuradhapura (East) at the Presidential Election. Inexamination-in-chief he was questioned as follows:-
20
Sri Lanka Law Reports
[1992] 2 Sri LR.
Q.Did you expect more votes than what you received at thePresidential election of 1987?
A. Yes.
Q.About how many votes did you expect?
A. I expected a minimum of another 20,000 more votes from myelectorate alone.
Q.For what reasons did you expect more votes?
A. After the defeat in 1977 we organized branches and formedwomen's organizations, youth organizations and enrolledmembers. We had more than 120 branches in my divisionalone and there were 8000 registered members and onemember from each household generally.
Q. Roughly, how many votes were there in support of the SLFP?
A. Two or three from each family.
Q. Were there any other reasons why you expected more votesto the SLFP?
A. In addition to the organizational strength there wasunpopularity on the part of the government party becausepeople could not get what they wanted; due to communalissues there were several killings near the JayasrimahaBodhi. This position was aggravated by the Indo-Sri LankaAccord.
Q. Are there any other reasons?
A. No.
Q. How many more votes did you expect in the entire district?
A. We expected at least another two lakhs more in theAnuradhapura district.
Similarly Pradeep Hapangama, M.P. for the Gampaha district testifiedto the “enthusiasm among the people for the SLFP in the Maharaelectorate”; that he enrolled 4000 members and established about100 SLFP party branches; that after the attack on the meetingorganized by the USA at Kadawata junction on 01.12.88 there was achange in the enthusiasm among the people to work for the SLFP.There was also the evidence of R. M. Jayasena, M.P. for theKurunegala district who stated that he formed 184 SLFP branchesand enrolled about 15300 members for the 1988 Presidential
$c
Sirimavo Bandaranalke v. Ranasinghe Premadasa and Another
(G. RS.de Silva, C.J.)
21
election. There was the evidence of witness Pina of Yapahuwa whoclaimed that of the 182 voters in his village all except 2 weremembers of the local branch of the SLFP of which he was thepresident. There is also the evidence of Dr. Neville Fernando, M.P. forthe Kalutara district. In examination-in-chief he stated as follows:-
Q. You said you expected a very high poll?
A. Yes, but they did not turn up.
Q. Is that the reason for the petitioner to get 28,000 votes?
A. Yes.
Q. Otherwise she would have got over 40,000 votes?
A. Yes.
Q. She received only 28,000 votes?
A. I expected that Mrs. Bandaranaike would get over 40,000votes.
Q. That was the result of this Intimidation caused by the shootingthat took place?
A. Yes.
This kind of evidence was led to show that in certain electoratesthe SLFP had strong support but that the acts of intimidationprevented the party supporters from votng on 19.12.88. Theinference to be drawn from the evidence is that but for the violencethe SLFP would have received more votes.
There is a further aspect to Mr. de Silva’s submission that the Courtis precluded from acting upon the evidence led on behalf of the 1strespondent. When Mr. Choksy moved to lead evidence on behalf ofthe 1st respondent at the conclusion of the evidence led on behalf ofthe petitioner, Mr. de Silva vehemently opposed it. It was thesubmission of Mr. de Silva that evidence of incidents outside what ispleaded in the petition is totally irrelevant to the issues that arise inthis case. Such evidence could, if at all, only strengthen the case forthe petitioner on the charge of general intimidation. Mr. de Silvaanalysed the constituent elements of s.91 (a) and we find thesubmissions he made then were substantially the same as thesubmissions he made before us in his closing address. On the otherhand, Mr. Choksy argued that he was entitled to lead evidence ofintimidation directed against the supporters of the UNP and his case
22
Sri Lanka Law Reports
[1992] 2 Sri LR.
was that such intimidation contributed to the low poll at thePresidential election. Upon a consideration of the submissions madeby Mr. de Silva and Mr. Choksy, this Court made order on 18.12.90overruling the objection taken on behalf of the petitioner. In allowingMr. Choksy to lead evidence of incidents outside the petition thisCourt made specific reference to the “order we have alreadydelivered in regard to the preliminary objection that was taken at thecommencement of these proceedings".
The order made by this Court on the preliminary objections isclearly binding on us, although Mr. de Silva argued that some of thecrucial findings therein are erroneous. Having regard to the nature ofthe preliminary objections that were raised, the Court was calledupon to analyse the ingredients of the charge of general intimidationpostulated in s. 91 (a). At the hearing on the preliminary objectionsthe foundation of the submissions of Mr. de Silva was that s. 91 (a)embodied the essential principles of the English Common Lawrelating to a free and fair election. The Court did not accept thiscontention. The Court ruled on what it considered to be the truemeaning of the words "… the majority of electors were or may havebeen prevented from electing the candidate whom they preferred".This ruling is undoubtedly a part of the ratio decidendi of the orderand it is not open to us to place a gloss on it or to deviate from it.What is more, we cannot overlook the significant fact that the trialproceeded on the basis of the interpretation placed by the Court ons. 91 (a) in the preliminary order.
One of the important contentions of Mr. de Silva both at the stagewhen he objected to evidence being led on behalf of the 1strespondent and also in his closing address was that any evidence ofan opinion expressed by a witness (be he an office-bearer of a localparty branch, a political activist, a local organizer or even an ordinarymember of a local branch) as to how others who are supporters ofthe party would vote is inadmissible for such opinion is based onpure conjecture, surmise and speculation. No one could predict howan elector who could not vote might have voted. In support of thisproposition Mr. de Silva relied strongly on a passage from thejudgment in Shiv Charan Singh v. Chandra Bhan Singh and Others™and on the observations of Grove, J. in the Hackney Case (,3>
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. RS.de Silva. C.J.)
23
which were cited in the preliminary order (1989 1 Sri L.R. 240 at 269and 270). The passage that was cited in the preliminary order fromShiv Charan Singh’s case reads as follows:-
“The burden to prove this material effect (on the result of theelection) is difficult and many times it is almost impossible toproduce the requisite proof. Electors exercise their right to voteon various unpredictable considerations, and the Courts are ill-equipped to speculate, guess or forecast by proceeding onprobabilities or drawing inferences regarding the conduct ofthousands of voters …". (1989 1 Sri L.R. at 269)
The observation of Grove J. in the Hackney Case cited in thepreliminary order reads thus:-
"I cannot see how the Tribunal can by any possibility say,what would or might have taken place under differentcircumstances. It seems to me a problem which the humanmind has not yet been able to solve, namely, if things had beendifferent at a certain period, what would have been the result ofthe concatenation of events upon the supposed change ofcircumstances …" (1989 1 Sri L.R. 270)
The point that is relevant and must be noted is that the Court in thepreliminary order cited these two passages when considering themeeting of the expression “were or may have been prevented” ins.91 (a). The court was at pains to point out the significant differencebetween the words “were” and “may have been". This is manifestfrom the reasoning of the Court which appears just before thequotation from Shiv Charan Singh’s Case. The Court reasonedthus:- “It seems to us that the term ‘may’ was designedly usedbecause mathematical proof that the majority of electors were in factprevented, in many a case, is impossible of attainment. The burdento prove that the majority of electors were in fact prevented is difficultand it is almost impossible to produce the requisite proof". (1989 1Sri L.R. 269) (1). We cannot agree with Mr. de Silva that there isanything in the preliminary order which precludes us fromconsidering the evidence led on behalf of the 1st respondent of acts
24
Sri Lanka Law Reports
[1992] 2 Sri LR.
of intimidation against his supporters. Indeed the Court overruled thisvery objection and allowed the 1st respondent to lead suchevidence. That order made in these very proceedings is clearlybinding on us. As stated earlier, both in Ratnam's case and Pelpola’scase the Supreme Court acted upon similar evidence. As submittedby Mr. Choksy, it was evidence of group leaders of voters that wasadduced by the petitioners and acted upon by the Court. And it wason the basis of that kind of evidence that the Court set aside theelection in these two cases.
Furthermore, in the preliminary order the Court held that one of theessential ingredients of the charge of “non-compliance” set out ins.91 (b) is that the “result of the election should be affected". At thepreliminary hearing Mr. de Silva argued the contrary, but in view ofthe ruling of the Court Counsel very properly conceded that hecannot maintain the “non-compliance" charge and abandoned it. Theresulting position is that in so far as a charge under s. 91 (b) isconcerned a Court must reach a finding as to whether the “non-compliance" affected the result of the election. A Court then mustconsider the question whether the petitioner would have succeededbut for “the non-compliance". For that purpose evidence of partyaffiliations would be relevant and admissible, notwithstanding thesecrecy provisions. Would it then be^ reasonable to say that thesecrecy provisions do not apply to s. 91 (b) but that they apply tos.91 (a)? We think not.
Before we conclude the discussion of the interpretation of s.91 (a)we think it fit and proper to observe that the construction sought to beplaced on s.91 (a) by Mr. de Silva is not without attraction. However,there is a definite ruling made by this Court in these sameproceedings, after a full argument, a ruling which runs counter to thesubmissions of Mr. de Silva as we have endeavoured to show. It isscarcely necessary to repeat that this ruling is clearly binding on us.
The facts in relation to the charge of “general intimidation” :•
We now turn to the facts as presented by the petitioner and the 1strespondent in respect of the charge of general intimidation. We willconsider the averments in the petition, the evidence led in support
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva, C.J.)
25
thereof by the petitioner, and the evidence adduced by the 1strespondent in respect of each electoral district as set out in thepetition.
Electoral District No. 01 – Colombo
Eleven incidents have been pleaded in the petition. The petitionerled evidence in regard to all incidents except the one pleaded inparagraph 7 (i) (g). It is unnecessary to burden this order with detailsof the evidence given by the several witnesses since Mr. Choksy didnot challenge these incidents. It is averred in paragraph (i) (a) that alarge number of voters at two polling stations were forcibly preventedfrom voting but the evidence does not bear out this allegation. Theother incidents related to bomb explosions at 4 polling stations; thechief SLFP organizer for Dehiwela was shot at on 17.12.88; explosionof bombs at the SLFP branch at Woodlands Mawatha in the Dehiwelapolling division on 19.12.88; the disruption of the meeting atBelekkade junction in the Ratmalana polling division on 16.12.88 bythe explosion of bombs when the petitioner was about to address themeeting; 3 persons were killed and 30 injured when bombs wereflung at a meeting held on 17.11.88 at Grandpass in support ofMr. Ossie Abeygunasekera.
Mr. Choksy stressed that the evidence did not disclose the killingof any SLFP organizer or supporter. There are 574 polling stations inthe Colombo electoral district but the evidence of incidents wasconfined to 7 polling stations.
As against this evidence, the 1st respondent led evidence ofnumerous killings, attacks, and threats of UNP organizers, office-bearers of party branches and of supporters in the following pollingdivisions:- Colombo Central, Colombo North. Homagama,Avissawella, Kolonnawa, Moratuwa, Kotte, Kesbewa, Kaduwela andMaharagama. A very large number of these incidents took place fromabout mid-September 1988. About 26 office-bearers of local partybranches were killed and 87 were attacked. A large number receivedthreatening letters and were compelled to put up bannersannouncing their resignation from the offices they held in the partybranches. They had no alternative but to refrain from engaging in any
26
Sri Lanka Law Reports
[1992] 2 Sri L.R.
political work. Several witnesses produced the threatening lettersthey had received. Some of the documents, purporting to be from theJVP, stated that the UNP was a party that was banned. (vide 1 R 42and 1 R 43) Some of the Jathika Sevaka Sangamaya (JSS)members who were active in the election campaign were killed andsome were attacked. It is in evidence that JSS members activelysupported the UNP in the election campaign. It is unnecessary toconsider in detail these anti-UNP incidents because the incidents assuch were not challenged in cross-examination.
Electoral District No. 02 – Gampaha
Paragraph (II) (a) of the petition speaks of an attack on theDharma Salawa polling station No. 27 in the Gampaha pollingdivision on 19.12.88 by a gang of unknown persons. This incidentwas not disputed by Mr. Choksy.
Paragraph (II) (b) alleges that unknown persons threw handbombs at the Ganegoda polling station in the Mirigama pollingdivision. There is no evidence to establish this incident. According toPolice Sergeant Hettiarachchi no incident took place at the pollingstation on the polling day. All that happened was that on the night of
while he was on patrol duty he heard a “loud sound" about50 yards away from the polling station and he fired in that direction.We accordingly hold that this incident has not been proved.
Paragraph (ii) (c) The incident averred is the throwing of bombs ata meeting held at Kadawata on 1st December 1988 in support ofMr. Ossie Abeygunasekera. Four persons were killed and someothers were injured. The incident is admitted. As a result of thisincident “a large number of people lost their enthusiasm to work forthe SLFP”.
There are 13 polling divisions in the Gampaha electoral district.Three incidents in 3 polling divisons are pleaded in the petition. Whilethere are 717 polling stations in the electoral district, the petitionercomplains of incidents only at 2 polling stations. There is noallegation of any incident directed against the SLFP. As far as the 'case for the petitioner is concerned, the evidence of generalintimidation in this electoral district is weak.
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P.S.de Silva, C.J.)
27
The case for the 1st respondent, however, stands on a differentfooting. L. P. Julis, a strong UNP supporter in Dompe who was tohave functioned as a polling agent on 19.12.88 was shot dead in hishouse on 18.12.88; Anthony Almeida, President JSS, Ja-Ela depot,was shot and killed on 14.11.88 at his residence; Hubert Silva, ChiefOrganizer, Ragama, and President UNP branch Ragama was shotdead on 18.11.88 at this house; S. M. Gunadasa, President JSS,Veyangoda Mills, was killed on 19.11.88. About 9 office-bearers ofUNP branches were killed from about mid-September 1988. Therewere as many as 133 resignations of members of the UNP branchesas well as from membership of the JSS on account of threatsreceived from the JVP. Some of these persons were compelled toannounce their resignations by inserting advertisements in thenewspapers – vide 1 R 57A, 1 R 57B, 1 R 52A, 1 R 55A, and 1 R 54A.
Mathew Perera a member of the Western Provincial Councilproduced the letter 1 R 58. This letter reads thus; “Resignation fromoffices”. “It is an act of treachery to serve the traitorous UNP -Thondaman Government that has betrayed the nation to the Indianimperialists … Therefore you are hereby ordered to resign forthwithfrom all the offices you hold. Penalty for defying this order is death.No further notice will be given”. There is a postscript which reads:'The UNP is a prohibited party. It is an act of treachery to carry ontheir propaganda activities…”.
Electoral District No. 03 – Kalutara
The petitioner has pleaded 10 instances of intimidation and hasled evidence in respect of 9. The petitioner has also pleaded postersthreatening voters in the Agalawatte polling division.
Paragraph (ill) (a) alleges that an SLMP supporter Rev. Premalokawas shot dead at Panadura on 19th December 1988. This incident isnot challenged.
The averment in paragraph (ill) (b) that a female voter who was onher way to the polling station on 19.12.88 was shot dead is notchallenged. Nor is the allegation in paragraph (iii) (c) that a voter atAgalawatte was shot and injured disputed.
28
Sri Lanka Law Reports
[1992] 2 Sri LR.
The incident pleaded in paragraph (iii) (d) is of a very seriousnature, and would undoubtedly have deterred many SLFPsupporters from going to the poll. It refers to the shooting incident atthe house of Dr. Neville Fernando on 18.12.88 at about 10.30 a.m.Eight persons were killed and several were injured consequent uponthe shooting.
The item pleaded in paragraph (iii) (e) reads as follows:- "A bombwas exploded near the Kaluwamodera polling station No. 43 inBeruwala polling division on 18th December 1988. Roads aroundpolling stations were blocked and Junior Presiding Officers andclerks did not report for duty*. We find that there is no evidence of abomb explosion near the Kaluwamodera polling station. Nor is thereevidence that Junior Presiding' Officers and clerks did not report forduty. We are satisfied that the incident has not been proved.
In regard to paragraph' (iii) (f) there is. evidence of a bombexplosion near the Andewela polling station No. 30 in the Matugamapolling division on 19th December 1988 at about 2 p.m. Voting wassuspended for about T 5 minutes and went on till 4 p.m. There is,however, no evidence that the explosion caused roads to beblocked.
Paragraph (iii) (g) alleges that the Senior Presiding Officer atGammana polling station No. 06 in the Agalawatte polling divisionwas shot at and injured in the early hours of the day of the poll, andthat several bombs were thrown near the polling station. Theevidence of the two witnesses called by the petitioner, namely,Seneviratne and Amerasena does not establish this allegation.
As against this the 1st respondent has led evidence of the killing ofabout 15 office-bearers of party branches in the Kalutara electoraldistrict from about mid-September 1988. There is the evidence of twostrong UNP supporters being killed at Agalawatte on 09.12.88. On
G. Wijesena, Secretary JSS Tibbotuwa branch was killed athis house at night. D. Pathirage, a political activist in Bulathsinghalawas shot dead on 10.12.88, as a bundle of UNP manifestos wasdiscovered in his house. H. de Silva, a member of the UNP branch atWaskaduwa North was shot dead on 21.11.88. As a result of this
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P.S.de Silva, C.J.)
29
killing all members of the Waskaduwa North party branch resignedand put up banners. Martin Vithanage, an active JSS member wasshot and stabbed to death on 08.11.88 in the Matugama pollingdivision. After the killing a poster was found by the road stating thatUNP traitors and stooges are punished with death. His widow andchildren left the village and did not vote. On 18.12.88 there was abomb explosion outside the house of Upali Wijekoon, a brother of thePresident of the local party branch. There were at that time 15-20UNP supporters inside the house. On 19.12.88 20 Tamil estateworkers on Delkeith estate in the Agalawatte polling division wereintimidated and they did not cast their votes. Anthony Cooray, Vice-President of the UNP Balamandalaya at Beruwela stated in evidencethat there were 300 voters at St. Vincent’s Home and 90% of themsupported'the UNP. At the Presidential election, however, only 6 voteswere cast.
Electoral District No. 04 – Kandy
In the Kandy electoral district there are 13 polling divisions and580 polling stations. The petitioner relies on one incident:- Shots werefired by unknown gunmen at voters who had gone to vote atDeliwalatenne. polling station (No. 24) in the Kundasale pollingdivision on 19th December 1988. The incident is admitted. Further itis not disputed that the voters were SLFP supporters and that theywere unable to cast their votes.
The examination-in-chief of the witness P. Karunadasa called bythe petitioner to speak to the incident is of relevance in view of Mr. deSilva's submission in regard to the provisions of law relating to thesecrecy of the ballet. Karunadasa was injured by the shooting.
Q. Have you been a supporter of any political party?
A. No. I am a voter of the Sri Lanka Freedom Party.
Q. You said you gave support by voting?
A. Yes.
Q. For which party?
A. It was Sri Lanka party.
Q. What party, what was the symbol?
A. Hand, The leader is Mrs. Bandaranaike.
30
Sri Lanka Law Reports
[1992] 2 Sri L.R.
Q. From when were you a supporter?
A. From 19651 am a supporter of that party.
Q. You said there were 6 persons who went along with you?
A. Yes, along with me.
Q. Did the others discuss with you where they were going?
A. We all were going to the polling booth.
Q. Did you know to which party they were supporting?
A. Yes, Sri Lanka Freedom Party.
In regard to the submission of Mr. de Silva that the party affiliation ofvoters is irrelevant, the evidence of witness Dambakotuwa (the SLFPorganizer for Kundasale) was that this shooting incident affectedvoters at other polling stations as well.
Q. After you visited Karunadasa at the hospital did you again goround your electorate?
A. Yes.
Q. You visited the polling stations?
A. Yes.
Q. What did you observe at these polling stations?
A. After getting the information of shooting at Delivalatennasupporters of the SLFP were frightened to come to the pollingstations to cast their vote.
Q. Was it only SLFP supporters who were frightened to comeand vote because of this incident?
A. Mostly.
It is not without significance that the shooting was at SLFPsupporters, and the evidence was that it affected mostly SLFPsupporters at the other polling stations as well. The extent of the"affectation" is seen from this evidence.
As against this single incident pleaded in the petition the 1strespondent has led evidence of the killing of about 11 office-bearersand of about 26 other supporters from about mid-September 1988.On 16.12.88 Sarath and Sisira Subawickrema (brothers) were takenout of their house and shot and killed. Sisira was the Secretary of thelocal UNP branch and Sarath was a member of the branch. On
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. RS.de Silva, CJ.)
31
one Dhanapala a strong UNP supporter was killed. Theseincidents occurred at the Galegedera polling division. On 07.11.88Nazeer Jamal, a UNP supporter was shot and killed at his boutiquenear the Kandy railway station. On 15.12.88 two organizers in twodifferent local areas in the Kundasale polling division (E. N.Gunawardena and D. R. Jayawardene) were killed. On the same daya UNP supporter named Kingsley Jayawardene was killed. On thenight of 18.12.88 3 persons named Najeem, Mani and Nisamdeenwho were putting up posters and distributing pamphlets for the UNPwere assaulted and had to be taken to hospital. On the same day
R.W. Vidurusinghe, Secretary of the branch at Putuhapuwa was shotand killed. In the Nawalapitiya polling division on 16.11.88 there wasan attempt to kill H. L. P. Tillakaratne, Chairman, Nawalapitiya UrbanCouncil and a staunch UNP worker. Three other UNP supporters whohad helped in constructing the stage for a public meeting on
at Nawalapitiya were shot and killed on 16.11.88.
It is in evidence that Tamil workers at Craighead and Monte Cristoestates (in the Nawalapitiya polling division) who were members ofthe CWC (and who were instructed by the Union tp vote for the UNP)were assaulted betweenlO and 11 a.m. on 19.12.88 and thusprevented from voting. In the Senkadagala polling division 5 personsengaged in putting up posters for the UNP were cut with swords andkilled on 18.12.88. H. E. Sumathipala, President of the UNP branch,Morayaya, in the Udu Dumbara polling division was shot and killedon 14.11.88. R. M. Gunathilleke, President of the UNP branch atTelagune in the same polling division was killed on 14.12.88. A UNPmeeting at Wattappola in the polling division of Udunuwara wasdisrupted on 13.12.88 and on the following day Kuda Banda, thePresident of the UNP branch at Wattappola was killed.
Kamala Randeniya (Secretary of the Kantha Samithiya,Muruthalawa branch in the Yatinuwara polling division) and herhusband received threatening letters directing them to resign from allposts held by them in the UNP. On 13.11.88 her brother was killed.Thereafter she and her husband put up posters in front of their house.One such poster dated 15.11.88 marked 1 R 71 was produced. Itreads thus:- “We Chandrapala Randeniya and Kamala Randeniya ofNo. 86 A, Muruthalawa hereby inform the patriotic comrades underoath that effective from 01.11.88 both of us have resigned from
32
Sri Lanka Law Reports
[1992] 2 Sri LR.
membership and other offices in the UNP and from offices in otherorganizations and that we will not participate in any political activitywhatsoever hereafter".
R. Abeysundera, M.P. for the Kandy district stated that duringthe 1988 Presidential election campaign there were numerous killingsof UNP supporters in the Yatinuwara polling division. Further therewere threats against persons working for the UNP. A number of themresigned from the party and refrained from election work. It was hisevidence that as a result of the threats the party activities in theelectorate came to a complete standstill. It was not possible to go outcanvassing and only a few supporters were willing to function aspolling agents.
Electoral District No. 05 – Matale
Matale electoral district has 4 polling divisions – Matale, Dambulla,Laggala, and Rattota. The total number of polling stations is 170. Thepetitioner has pleaded 11 incidents and has led evidence on 10incidents. The petitioner has pleaded threatening posters in all 4polling divisions.
Paragraph (v) (a) This relates to the attack on the house of theSLFP organizer for Rattota on the night of the 18th of December1988. This was an incident of a very grave nature. Six persons in thehouse were killed including the SLFP organizer Wegodapola. Theincident was not challenged. It would undoubtedly have frightenedvoters, particularly the SLFP supporters whose organizer was killedon the eve of polling day.
Paragraph (v) (b) The Madawala Ulpotha polling station wasattacked on 19.12.88 while voting was going on and 4 voters werekilled. The SPO closed the poll by 12 noon owing to the shooting. Theincident is admitted but evidence was led on behalf of the 1strespondent that the voters killed were UNP supporters.
Paragraph (v) (c) No evidence was led in support of theallegation that the Galewela polling station in Dambulla pollingdivision was attacked.
sc
Sirfmavo Bandaranaike v. Ranasinghe Premadasa and Another
(G.P.S.de Silva, C.J.)
33
Paragraph (v) (d) The allegation here in that the Kalundawepolling station (No. 40) in the Dambulla polling division was attackedon 19.12.88 while voting was going on and that 2 voters were injured.There is no evidence that the polling station was attacked – vide theevidence of V. G. Wijekoon Banda. No officer who functioned at thepolling station was called. We hold that this incident has not beenproved.
Paragraph (v) (e) It is alleged that the Elamalpotha polling station(No.18) in the Dambulla polling division was attacked on 19.12.88.The evidence does not support this allegation – vide the evidence of
S.M. Imamdeen who admitted that the polling station was notattacked. We accordingly hold that this incident has not beenestablished.
Paragraph (v) (f) The murder on 17.12.88 of the SLFP organizerfor Dambulla, T. B. Kulatunga, was not challenged. Mr. Choksysubmitted that Kulatunga’s office was a very small one, consisting ofonly 2 rooms, one of which was used as a boutique. Counsel alsostressed that the office was situated by a gravel road inGalahitiyagama which is a small village on the border of Dambullapolling division. These circumstances do not, however, detract fromthe important fact that Kulatunga was the SLFP organizer forDambulla and his murder on the eve of the election wouldundoubtedly have driven fear into the minds of the voters, particularlythe SLFP supporters. At one stage of the cross-examination, the factthat Kulatunga was the organizer was challenged but later this wasnot pursued.
Paragraph (v) (g) The burning of the office of the SLFP atWalawela on 13.12.88 was not disputed. This would have had anadverse effect on the SLFP campaign.
Paragraph (v) (h) The throwing of bombs at the SLFP main officein the Matale town on 16.12.88 at 5 a.m. was also not challenged.This too was an incident which would have affected SLFP supporters.
Paragraph (v) (i) The shooting on 09.12.88 of the President of theSLFP branch at Aluthgama, is not challenged. This incident wouldhave had an adverse impact on the SLFP supporters.
34
Sri Lanka Law Reports
[1992] 2 Sri L.R.
Paragraph (v) (j) The averment here is that:- “posters appearedbefore the election day warning people not to vote for the SLFP”. Theevidence of C. W. Abeyratne shows that in Madipola in the Dambullapolling division posters appeared stating “death for supporters of theSLFP". The witness Chandawimala Thera said he saw posters whichread "Refrain from voting for the SLFP" in 4 places in the Matalepolling division. There is also the evidence of Haniffa Hadjiar to thesame effect. We hold that the averment in paragraph (j) has beenproved.
Paragraph (v) (k) It is averred here that gangs of unknownpersons went from house to house warning persons not to vote. Theevidence of O. W. Abeyratne, Upatissa Banda, D. D. Kannangaraand Chandawimala Thera establishes this fact.
On a consideration of the above evidence we hold that there wasconsiderable intimidation directed at the SLFP supporters in theMatale electoral district as a result of the killing of 2 SLFP organizers,the attacks on 2 SLFP offices, and the posters specifically warningpeople not to vote for the SLFP.
The 1st respondent led evidence of resignations of office-bearersin the local party branches and of a very large number of JSSmembers who were compelled to resign on account of death threats.District Secretary of the JSS (Central Province) Sydney de Soysa saidin his evidence that there were about 2500 workers in the 6 C.T.B.depots that fall within the district and that 2150 are members of theJSS. Premasiri, the President of the JSS, Matale depot was shot at inNovember 1988 and the result was that the election work done by theJSS came to a standstill. Practically all the office-bearers of the 6depots resigned from their posts. He was questioned as to the kind ofpolitical work done by the JSS members during elections. His answerwas, “They assist in fixing posters, if there is to be a rally, they help toconstruct the stage; we organize big meetings. We do in shortpropaganda work". But at the Presidential election the JSS memberswere unable to engage in such activities.
The evidence further shows that about 10 office-bearers in partybranches and about 15 other supporters were killed from about
sc
Sirimavo Bandaranaike v. Ftanaslnghe Premadasa and Another
(G. P. S. de Silva, C.J.)
35 ■
mid-September 1988 in the Matale electoral district. On 18.11.88
K.G. Loku Banda a staunch party supporter in the Matale pollingdivision was killed. The treasurer of the Kotuwegedera branch waskilled on 22.11.88 and a member of the same branch was shot at inthe course of the same incident. A prominent UNP workerM. J. Shahabdeen from Madipola was killed on 17.12.88. On thesame day A. L. M. Sarada, chief organizer of Undugoda PalleSiyapattuwa was attacked.
There was evidence of a somewhat unusual incident that tookplace on 17.12.88 in the Laggala polling division. Biso Menika, astrong UNP supporter was threatened, her hair was cut and wasordered to carry a poster and cycle a distance of 15 miles up toHettipola and back to her village, passing through 7 villages. Sheproduced the posters (1 R 81 and 1 R 82) and also produced the cuthair in court. The poster reads, ‘This is the punishment for findingfault with the patriotic comrades”. By the 15th of December most ofthe UNP chief organizers of the Laggala polling division had resignedowing to death threats.
It is unnecessary to enumerate the other incidents relating to thekillings of, and attacks on office-bearers and party supporters. Thereis little doubt that the intimidation directed against the supporters ofthe 1st respondent was by no means insignificant.
Electoral District No. 06 – Nuwara Eliya
Paragraph (vl) (a) reads thus:- "A large number of voters ofThibbotugoda polling station (No. 21). Rupaha polling station(No.44) in the Walapone polling division were forcibly prevented fromvoting for the candidate of their choice by gangs of unknown personswho had blocked the access roads to the polling stations”. There isthe evidence of T. B. Wickremasinghe that a water channel had been 'diverted across the access road to Thibbotugoda polling station.There were 552 registered voters but only 2 votes had been cast. Asregards Rupaha polling station there is no satisfactory evidence toestablish that the access road was blocked.
Paragraph (vi) (b) Although the allegation here is that the housesof SLFP supporters were burnt in Beramana village on 04.12.88, the
36
Sri Lanka Law Reports
[1992] 2 Sri LR.
evidence of the Police officer establishes that only one house of anSLFP supporter was burnt on 07.12.88. This is not challenged.
There are 272 polling stations in the electoral district but thepetition refers to incidents in only 2 polling stations. There is nocomplaint in respect of threatening posters. The voter turn-out at thePresidential election was as high as 79.96%.
Electoral District No. 07 – Galle
Paragraph (IvA) (a) states that for 2 weeks prior to 19.12.88unknown persons threatened A. M. Karunaratne, SLFP organizer forAmbalangoda and his wife at their house. This is clearly proved bythe evidence of Karunaratne and his wife. In fact the incident isadmitted.
Paragraph (IvA) (b) alleges that the house of Saman de Silva co-organizer for Ambalangoda was burnt. This incident too is notchallenged.
There is evidence of threatening posters in the polling divisions ofAmbalangoda, Balapitiya, Bentota-Elpitiya, Karandeniya andRatgama.
On the other hand the evidence led on behalf of the 1strespondent shows that about 40 office-bearers of party brancheswere killed from mid-September 1988. B. G. Bandusena, Secretary,Doralla branch, Osmund Jayasooriya, Secretary Youth League,Mahawatte, S. Wimalasooriya, Treasurer, Patabendimulla branch,
K. Y. Lokuge, Secretary Mlukpitiya branch, P. V. Piyadasa,President JSS branch at the Plywood Corporation, Gintota, were allkilled in the months of October, November and December 1988. Inmany instances a poster was found by the body of the deceasedstating that the reason for the killing was working for the UNP. In theBalapitiya polling division 5 strong UNP supporters were tied togetherand killed on 19.11.88. M. M. Nandasena, Secretary UNP branchNugaduwa in the Akmeemana polling division was killed on 09.11.88.In the Ambalangoda polling division alone 16 office-bearers andsupporters were killed between 23.10.88 and 12.12.88. Consequent
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva. C.J.)
37
upon these killings several local organizers and office-bearersresigned from their posts by displaying banners and putting upposters. The election campaign and the organisation suffered aserious setback. In addition there is evidence of posters specificallydirected against the UNP and its supporters in the Ambalangoda,Bentara-Elpitiya, Habaraduwa, Galle, Baddegama, Karandeniya,Baiapitiya and Hiniduma polling divisions.
Electoral District No. 08 – Matara
Paragraph (vli) (a) The petitioner relies on one incident, viz. alarge number of voters of Buddha Jayanthi polling station (No. 08) ofHakmana polling division were prevented from voting for thecandidate of their choice by gangs of unknown persons who hadblocked the access roads to the polling station. The petitioner called2 witnesses (Amaradask and Sethupala) but their evidence iscompletely contrary to the allegation in the petition. According to theevidence of the Senior Presiding Officer Amaradasa, he reached thepolling station and the road had been cleared. We acordingly holdthat the allegation has not been proved.
The petitioner led evidence of intimidatory posters in the pollingdivisions of Hakmana, Akuressa, Deniyaya, Weligama, Devinuwaraand Kamburupitiya. There is one witness, Sumanawathie Pahalage,of Akuressa called by the petitioner to give evidence on posters whoadmitted in re-examination that the killings affected mostly the UNPand its supporters.
The Matara electoral district consists of 7 polling divisions andthere are 358 polling stations. The 1st respondent has led evidenceof killings and attacks on his supporters in the polling divisions ofKamburupitiya, Akuressa, Deniyaya, Weligama, Devinuwara andHakmana. The evidence shows that 53,office-bearers of UNPbranches and other supporters were killed between September andDecember 1988. H. H. Sirisena, President of Ududamana partybranch in the Kamburupitiya polling division was shot and killed on
and his head was severed from the body. K. Siyaneris,organizer for the UNP at Diganahena in Akuressa polling division wasshot and killed on 15.10.88. M. L. Diyonis, the organizer for
38
Sri Lanka Law Reports
[1992] 2Sri L.R.
Pahala Maliyaduwa was shot dead on 18.09.88. R. Vettasinghe,Secretary of the Malimbada Youth League was shot dead on
A poster put up after his death stated that he was killed bythe JVP because he supported the UNP. On 12.12.88K. G. W. Rajapaksa, Superintendent of Wilpita State Plantation and astaunch UNP supporter was killed in his office. A. R. Siriwardena, anactive UNP worker and a member of the JSS who worked during thePresidential election campaign was stabbed to death on 17.12.88.After his death, posters appeared stating that he was killed becausehe was a staunch UNP supporter. On the same day another activeUNP worker was killed and posters appeared later giving the reasonfor the killing – that he was a supporter of the UNP. FrancisWeeraman, another strong UNP supporter was shot and killed on
After his death posters appeared stating "death is thepunishment for traitors who worked for the UNP”. It is in evidence thatalmost all the office-bearers of the JSS branch of Matara C.T.B. depotresigned in October 1988 consequent upon receiving threateningletters. On 09.12.88 R. P. Gamini, Secretary of the Urubokka BalaMandalaya and an undergraduate of the University of Ruhuna waskilled.
V. P. Abeywickrema, a member of the Provincial Council for theSouthern Province stated in his evidence that the numerous threatsand killings of UNP supporters seriously affected the electioncampaign and organisational work of the UNP. He further said “allthis affected the voting of the UNP supporters at the election". D. A.Wickremasinghe, M.P. for the Matara District also testified to theeffect the attacks and the killings had on the election campaign:-
Q. At the time of this Presidential election campaign of 1988what was the state of the UNP organizations at village level?
A. Our organization had got paralysed at village level becausesome office-bearers had been killed and some had left thevillage.
Q. As a result what was the effect of this on the Presidentialelection campaign of 1988 of the UNP?
A. Our organisational capacity became very much weakened.
Q. Were you able to hold meetings at village level?
A. No…
sc
Sitimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva, C.J.)
39
Q. Were you able to go house to house canvassing?
A. No.
Q. The normal election work could not be done?
A. That is so.
Q. Were you able to get polling agents?
A. No, we were unable to get.
Electoral District No. 09- Hambantota
Paragraph (vlii) (a) reads thus:- “SLFP organizer (R. Dharmasena)who was in charge of 27 polling stations in the Mulkirigala pollingdivision was shot dead by unknown persons on 15.12.88". Thisincident has not been challenged. The widow Kusumawathie spoketo the circumstances in which her husband came by his death.
Paragraph (vili) (b) The incident pleaded here is that anotherSLFP organizer, Dissanayake, whb was in charge of 15 pollingstations in the Mulkirigala polling division was killed by unknownpersons before the elections. This killing too is admitted. It wouldappear that he was killed on the night of 14th December 1988.
The killing of the SLFP organizers in the Mulkirigala polling divisionwould undoubtedly have had a grave impact on the supporters of theSLFP. The petitioner has also led evidence of intimidatory posters inall the 4 polling divisions in the Hambantota electoral district.
The 1st respondent led evidence of threats, attacks, and a fewkillings of office-bearers and party supporters during the Presidentialelection campaign. Harry Abeydeera, M.P. for the Hambantotadistrict testified to the effect of those ingidents:-
Q. And for the Presidential election in 1988 could your partyorganization in Beliatte electorate function at all?
A. No.
Q. Was your UNP organization in the Tissa electorate also ableto function for this Presidential election in 1988?
A. No.
Q. In Mulkirigala?
A. No.
Q. And in Tangalla too was the position the same?
A. Yes.
40
Sri Lanka Law Reports
[1992] 2 Sri L.R.
He also referred to the meetings held in support of the 1strespondent at Katuwana on 12.11.88, at Beliatte on 20.11.88 and atTangalla on 27.11.88, all of which proved to be a failure owing tobomb explosions and shots being fired in the vicinity of the venue ofthe meetings. At the meeting at Tangalla posters appeared warningpeople not to attend the meeting. UNP polling agents too were afraidto function at the polling stations in the Hambantota electoral district.
Chamal Rajapakse, M.P. for the Hambantota district called by thepetitioner admitted that persons from families who were known to besupporters of the UNP were killed in Walasmulle and Middeniya. Thiswas between 10th November and 19th December 1988. When askedwhether the UNP "suffered from the killings of its party supportersand workers’, his answer was “that started with the signing of theAccord".
Electoral District No. 15 – Kurunegala
The Kurunegala electoral district has 14 polling divisions and 638polling stations. At the 1988 Presidential election the voter turn-out forthe Kurunegala electoral district was 50.05%. Apart from theincidents set out in the petition, intimidatory posters have beenpleaded in 4 polling divisions.
Paragraph (xl) (a) The averment here is that the access roads tothe Kudagalgamuwa polling station (No. 10) were blocked and voterswere harassed by gangs of unknown persons.
The petitioner called 5 witnesses. Two of these witnesses(R. M. Tilakaratne and R. M. Punchi Banda) remained at Minhettiyawhich is about 3 miles away from Kudagalgamuwa and could nothave known whether the access roads were blocked or not. Witness
S.B. M. Ranasinghe had walked from Edandawela toKudagalgamuwa and denies the existence of road blocks. The onlywitness who speaks to road blocks is T. M. Ranbanda. It is thus seenthat the evidence is of a contradictory nature. We accordingly holdthat the allegation has not been proved.
Paragraph (xi) (b) The petitioner led no evidence in regard to thealleged attack on polling station Mahakeliya (No. 06) in Wariyapolapolling division.
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva, C.J.)
41
Paragraph (xi) (c) Polling station Hewanpellessa (No. 25) in theNikaweratiya polling division was attacked on 19.12.88. This incidenthas not been challenged.
Paragraph (xl) (d) The allegation is that persons who came to voteat polling station Yayawatte (No. 25) and Galpola (No. 30) were shotat and two voters were killed. The evidence establishes that one voternamed Najibudeen was killed at polling station Yayawatte (No. 25).No evidence was led in regard to any incident at polling stationGalpola (No. 30).
Paragraph (xi) (e) It is averred that persons who came to vote atpolling station lhala Gomugonuwa (No. 29) were shot at on 19.12.88and one person died. This incident is admitted and the person whodied is Tikiri Banda, a supporter of the SLFP. The incident occurredduring polling hours.
Paragraph (xi) (f) On 18.12.88 two SLFP supporters ofNikeweratiya were killed by unknown persons. This incident was notchallenged.
Paragraph (xi) (g) On 18.12.88 the house of A. Tennekoon, theSLFP organizer for Nikaweratiya was attacked in the night byunknown persons who threatened him and his family. This incidenttoo was not disputed.
Paragraph (xi) (h) Tour members of the SLFP in the Galgamuwapolling division were brutally killed just prior to the elections". Twowitnesses, S. H. Podiratne and Bandula Basnayake were called tospeak to this incident. The evidence clearly proves that 4 personswere killed and 3 of them were SLFP supporters.
Paragraph (xi) (i) “A large number of voters in Galgamuwa pollingdivision were threatened by gangs of persons by shooting and otherforms of intimidation". There was only one witness, BandulaBasanayake SLFP organizer for Galgamuwa called to testify to thisincident. He spoke of a gang of persons attacking his vehicle whenpassing through a jungle area on his way to a polling station onpolling day. There is no evidence to prove the allegation as pleaded
42
Sri Lanka Law Reports
[1992] 2 Sri LR.
in the petition. We accordingly hold that the incident has not beenestablished.
Paragraph (xl) (j) “A large number of voters in the Hamangallaarea in the Katugampola polling division were threatened by gangs ofunknown persons coming to their homes and threatening them not tovote, firing shots in the air in the night as well as putting up postersthreatening them not to vote". The petitioner called 3 witnesses whoasserted that gangs of persons visited their homes and threatenedthem not to vote. Two of these witnesses however did function aspolling agents. We hold that the evidence establishes that voterswere threatened between the 15th and 17th of December 1988 not toexercise their vote.
On a consideration of the evidence led on behalf of the petitionerwe hold that there was a fair degree of intimidation directed at theSLFP supporters in the Kurunegala electoral district.
As against this, Mr. Choksy for the 1st respondent led a substantialvolume of evidence of killings and attacks on organizers, office-bearers and supporters of the UNP. The evidence reveals that therewere 62 killings of UNP office-bearers and supporters in the monthsof September, October, November and December 1988. (up to19.12.88) in the Kurunegala electoral district. These incidents were ofa widespread nature covering 10 out of the 14 polling divisions of theKurunegala electoral district. Lionel Jayatilaka, a Cabinet Minister,was killed on 26.9.88 and his death frightened office-bearers andmembers of party branches in Kuliyapitiya. J. S. Jane Nona, Chieforganizer, UNP Women’s League in Kuliyapitiya was shot at on
and thereafter she left the village. J. A. Nilangaratne, thetreasurer of the local party branch and an active worker was killed on
A. W. M. Thaha, President of the local party branch and theleader of the Muslims in the village of Arakyala was killed on
D. M. Abeyratne Banda, the organizer of theMedagamdahaya Korale Bala Mandalaya was shot and killed on
This resulted in office-bearers giving up party work throughfear. Three UNP supporters in the Dambadeniya polling division werekilled in the same incident on 11.12.88 and the killers had left aposter which stated that they were killed because they worked for the
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva, C.J.)
43
UNP in their village. In the Dambadeniya polling division severaloffice-bearers who had resigned from their posts in their partybranches had displayed banners announcing their resignation. Thebrother of M. H. B. Wanninayake, M.P. was killed on 12.11.88 in theNikeweratiya polling division. H. M. A. Loku Banda, M.P. for theKurunegala district stated that he saw posters in the Galgamuwapolling division saying that the UNP supporters are prohibited fromvoting. J. Balasuriya, President of the Koswatte party branch in thePanduwasnuwara polling division was killed on 25.11.88. There isevidence to show that there were posters specifically directed at theUNP in the polling divisions of Kuliyapitiya, Wariyapola, Nikaweratiya,Katugampola, Panduwasnuwara, Yapahuwa and Dodangaslanda.
Electoral District No. 16 – Puttalam
The petitioner has pleaded 4 incidents but no evidence was led inregard to two incidents, namely, paragraph (xii) (c) and (d). Evidencewas led in respect of paragraphs (a) and (b).
Paragraph (xii) (a) “In the Anamaduwa polling division aVen. Buddhist monk who is an SLFP supporter was dragged out ofhis temple in Wadigamangawa and mercilessly assaulted”. WitnessKarunaratne speaks to the assault pn the priest but he says theassault appears to have been on account of his personal conductand not due to any political activity. We hold that the incident aspleaded has not been proved.
Paragraph (xii) (b):- "Shooting at and intimidation of voters tookplace at the polling station No. 13 (Thalgaswewa) by unknownpersons”. Wimalasuriya was the witness called to speak to thisincident. He said that he heard gunshots around the polling station.No officer who functioned at the polling station was called by thepetitioner. There is no record of the-alleged shooting in the journal.The 1st respondent called a police officer who was on duty at thepolling station and he denied that such an incident took place. Theevidence is unsatisfactory and we hold that the alleged incident hasnot been established.
On the other hand, the 1st respondent led evidence of the killing of3 office-bearers of party branches in the Puttalam electoral district, in
44
Sri Lanka Law Reports
[1992] 2 Sri L.R.
the month of December 1988. Keerthi Sovis, President of the UNPYouth League, Mahuswewa, was shot dead on 16.12.88. There was aposter beside his body which stated: “This is the punishment fortreacherous UNP supporters”. Sunil Ananda, Secretary of the localparty branch was shot and killed on 17.12.88. A poster was put upstating “Sunil Ananda was killed because he was a stooge of theUNP"; I. B. Fernando, President of the UNP Youth League atWennappuwa and President of the party branch at Waikkal was killedon 02.12.88.
Electoral District No. 17- Anuradhapura
Paragraph (xlii) (a) “On 19.12.88 two voters H. F. Mohammed andCarim of Mookiriyawa in the Horawapathana polling division wereshot dead by unknown persons". Three witnesses namely,Shahabdeen, Piyadasa and Inspector of Police Ratnayake werecalled to testify to this incident. The incident is admitted. However,the 1st respondent called a witness named N. Cader who claimedthat the two deceased persons were UNP polling agents, killed ontheir way to the polling station. This belated claim is not acceptableas it was never put to the witnesses called by the petitioner.
Paragraph (xiii) (b) “On 19.12.88 election staff travelling to thepolling station at Parangiya Wadiya Ranpathwila in a vehicle wereshot at by unknown persons and one Piyadasa was injured in theHorawapathana polling division”. The incident is not disputed.Mr. Choksy pointed out that there is no complaint that the pollingstation did not function on account of this attack.
Paragraph (xiii) (c) “On 19.12.88 election staff travelling invehicle No. 29 Sri 247 was shot at by unknown persons". Theincident is admitted. There is evidence to show that the poll was infact conducted at the polling station. No one was injured.
Paragraph (xiii) (d) “On 19.12.88 police mobile patrol of S. I.Wijekoon was shot at and attacked. S. I. Wijekoon, the driver and aHome Guard were injured and one Army Private died. This happenedin the Medawachchiya polling division”. This incident is admitted. Itoccurred when the vehicle was proceeding in a jungle area.
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G.P.S.de Silva. C.J.)
45
Paragraph (xiii) (e) “In Kalawewa polling division in the pollingstation area No. 24 (Katiyaya Yaya) 5 SLFP supporters were killedabout 15 days prior to the election day by unknown persons. Twelvedays prior to the election day another SLFP supporter was killed andon the day before the election another supporter of the SLFP waskilled by unknown persons”. The killings are admitted. The allegationthat those killed were SLFP supporters is disputed. The evidence ofMuthukumarena is sufficient to prove that the 5 persons killed about15 days prior to the election were SLFP supporters. However, theidentity of the other 2 persons killed has not been established.
Paragraph (xiil) (f) “In the polling division of Mihintale unknownpersons armed with guns prevented voters from voting at pollingstation Manewa (No. 22)”. The evidence led on behalf of thepetitioner does not prove the incident as pleaded. The chief witnesswas M. Ariyadasa. All that the evidence establishes is that Ariyadasaalone was prevented from proceeding to the polling station by anarmed gang of persons who assaulted him. Ariyadasa was an SLFPpolling agent.
Paragraph (xiii) (g) "On 19.12.88 polling station No.26(Kongahawewa) was attacked by unknown persons”. No evidencewas led to prove this allegation.
The petitioner also led evidence of threatening posters in thepolling divisions of Horawapathana, Kekirawa and Mihintale.
The 1st respondent called evidence to establish killings of, andattacks on, organizers, office-bearers and supporters of the UNP inall 7 polling divisions of the Anuradhapura electoral district. Theevidence shows that about 39 office-bearers and supporters of theUNP were killed during the period September to December 1988. Inthe Anuradhapura district there were 7 C.T.B. depots and each hadits own JSS branch. On account of threats by the JVP almost all theoffice-bearers in the JSS resigned. As ordered by the JVP they put upbanners stating that they have resigned from the UNP and the JSS.Witness A. Jayatilake, a UNP supporter in Kekirawa polling divisionsaid that on 15.12.88 he saw posters put up opposite his housestating "UNP voters will be killed”. A. B. Ariyadasa, President of the
46
Sri Lanka Law Reports
[1992] 2 Sri LR.
UNP Bala Mandalaya, Kalawewa, was killed on 24.11.88. There wasa writing near his body which stated that he was killed for working forthe UNP. T. M. Abeyratne, an active UNP supporter who addresedseveral meetings in Anuradhapura East and the adjoining electorateswas shot at and injured on 21.11.88. Earlier he had receivedthreatening letters from the JVP. D. Kithsiri, President of the UNPbranch, Turuwila, was shot dead in his house on 03.12.88. His familywere supporters of the UNP but after his death they refrained frompolitical work. M. B. Basnayake, President of the UNP branchOttappuwa was killed on 07.12.88. A poster was found near his bodystating that death is the punishment for working for the UNP. Theevidence also shows that a very large number of office-bearers in theUNP branches resigned and kept away from political work owing tothreats from the JVP.
Electoral District No.18 – Polonnaruwa
Paragraph (xiv) (a) “In the polling division of Medirigiriya at pollingbooth No. 09, Thelawewa, unknown persons opened fire and injured5 persons on 19.12.88”. This incident is admitted but the number ofpersons injured is 3 and not 5, as is seen from the evidence of thepolice officers called by the petitioner. Moreover, the 1st respondentcalled B.K. Guneratne who asserted that the 3 injured persons wereUNP supporters. No evidence to the contrary was led by thepetitioner.
Paragraph (xiv) (b) “In the polling division of Polonnaruwa aperson by the name of J.M. Jayawardena was shot dead by unknownpersons on 19.12.88". This incident is not disputed. However, theevidence of Police Inspector Gunasekera called by the petitioner andthe evidence of B.K. Gunaratne called by the 1st respondent showthat the deceased was a UNP supporter.
Paragraph (xiv) (c) “In the polling division of Polonnaruwa aperson by the name of Ariyasena was shot dead on 19.12.88.” Theincident is admitted. The victim was a supporter of the UNPaccording to the evidence of Police Inspector de Silva called by thepetitioner and also the evidence of B. K. Guneratne called by the 1strespondent.
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P.S.de Silva. C.J.)
47
It would thus appear that on the evidence all 5 victims in thePolonnaruwa electoral district were supporters of the UNP. Even inthe pleadings the petitioner did not claim that they were supporters ofthe SLFP.
As against this, the evidence led on behalf of the 1st respondentreveals that 13 UNP office-bearers and supporters were killedbetween September and December 1988. W. C. Boyagoda, SecretaryUNP branch at Hathamune was shot dead on 23.09.88. QuintusFernando, President of the UNP branch of Henyaya and member ofthe Bala Mandalaya was shot and killed in his house on 05.12.88.S. A. Maithripala, Secretary of the UNP branch at Yatigalpatana waskilled on 17.12.88.
Electoral District No.19- Badulla
Paragraph (xv) (a) "In the polling division of Welimada atWeegolla a Senior Presiding Officer was shot dead and a civilian anda police constable were injured on 19.12.88”. The incident isadmitted. It took place at about 5.10 p.m. after the close of the pollwhen the vehicle was returning to the Badulla Kachcheri. The ballotboxes were not damaged and were taken to the Badulla Kachcherifor the count. This incident which took place after the poll had noadverse effect on voters.
Paragraph (xv) (b) "In the polling division of Welimada booth No.30 (Alugolla) and No. 40 (Ohiya) were attacked by unknownpersons”. The attack on booth No. 30 (Alugolla) is admitted, but thereis no evidence of an attack on booth No. 40 (Ohiya). The attack tookplace at 12.45 p.m. and at that time there were only 3 voters in thequeue and they had run away.
Paragraph (xv) (c) "In the polling division of Bandarawela policefound 8 persons murdered at Ellethota near the railway bridge on19.12.88". This incident is not challenged. It was elicited in cross-examination from Police Inspector Jayatissa that all the deceasedpersons were supporters of the UNP. However, the incident occurredwell after the close of the poll and it could not have had any adverseeffect on voters.
48
Sri Lanka Law Reports
[1992] 2 Sri LR.
Paragraph (xv) (d) "Between 3rd and 5th December 1988 in thepolling division of Viyaluwa unknown persons in police uniform atMeegahakivula threatened SLFP supporters not to vote on 19.12.88”.This incident is not challenged. The evidence indicates that therewere houses of UNP supporters also in this village.
Paragraph (xv) (e) In the Bandarawela polling division a bombwas thrown at W. Ratnayake, Chief organizer’s (SLFP) residenceNo. 20, Badulla Road, Bandarawela on 18.12.88”. This incident is notchallenged.
Paragraph (xv) (f) “The SLFP chief organizer for Passara pollingdivision, D. G. M. Landawela was shot dead on 17.12.88 by unknownpersons". This incident is admitted. It would have definitely affectedthe SLFP supporters adversely in the Passara polling division.
The evidence led on behalf of the 1st respondent shows thatduring the period September to December 1988 as many as 18office-bearers of the party branches were killed in the Badullaelectoral district. W. M. Amerasekera, President of the UNP branch atBoralanda and a member of the UNP Youth League was killed on
In the same incident a supporter of the UNP called
R.M. Jayasena was also killed. R. M. Razak, the UNP organizer forWelimada and Uva Paranagama received a number of threateningletters. He thereafter stopped working for the UNP and left his villageten days prior to 19.12.88. L. M. Muthu Banda, President of theRilpola Korale UNP branch who had earlier received threateningletters was shot dead on 27.11.88. David Appuhamy, SecretaryBogoda North Bala Mandalaya was killed on 11.12.88. R. M. B.Ratnayake, President of the UNP Youth League, Katugaha branchand President of Deluwina Korale Balamandalaya resigned from theoffices he held in November 1988 by publishing a notice anddistributing it in the village – vide 1R 67. A. J. M. Upasena, Presidentof the Dowa UNP branch and JSS member, Bandarawela, stated thatin October and November 1988 three active JSS members werekilled. It is in evidence that the JSS worked actively for the UNP atevery previous election.
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva, C.J.)
49
Electoral District No. 20 – Moneragala
Paragraph (xvi) (a) “On 18th December 1988 (unknown persQnsin uniform) removed SLFP posters and the SLFP office board inMoneragala”. This incident is not admitted by Mr. Choksy. Thepetitioner called H. M. Wijeratne to speak to the incident. He waspresent on the occasion the board was removed. He admitted that hemade no complaint to the police and the reason was that he wasafraid to do so. We accept his evidence and we hold that the incidenthas been proved.
Paragraph (xvl) (b) “Trees were cut down and the road obstructedfrom Badalkumbura to Passara. Four culverts were damaged and theroad from Hingurukaduwa to Badalkumbura was made impassableon 19.12.88”. Two witnesses were called by the petitioner, namely,Sugathadasa and Wijekoon. Admittedly Sugathadasa did not go outof his house on 17th, 18th and 19th December and he could not haveknown anything about the state of the road. Wijekoon claimed that hesaw the obstructions on the road but he further stated that he had asuspicion that the posters appearing at the time of the Presidentialelection were put up by the UNP and that the UNP may have killedtheir own supporters. His evidence when considered as a whole isunacceptable. We hold that the allegation is not proved.
Paragraph (xvi) (c) "In the Wellawaya polling division when SeniorPresiding Officer S. Abeysundera and his staff were proceeding tothe polling station (No. 22) Tanamalwila his vehicle was stopped atBaddandiyaya and fired at, Two police officers sustained injuries”.The incident is admitted. It occurred on 18th December 1988 atabout 8 p.m. There is no evidence that the polling station did notfunction on 19.12.88.
Paragraph (xvl) (d) “In the polling division of Wellawaya mobilepolice, patrol was attacked on 19.12.88 by unknown persons and 3police constables were injured”. No evidence was led in support ofthis allegation.
Paragraph (xvi) (e) “In the Moneragala polling division 2 Armysoldiers were shot at and injured by unknown persons on 19.12.88”.
50
Sri Lanka Law Reports
[1992] 2 Sri LR.
The incident is admitted. The evidence shows that the election stafftravelled in another vehicle and they proceeded to the polling station.The conduct of the poll was not affected by this incident.
Paragraph (xvl) (f) “In the Moneragala polling division, S. I. Silvaand party while on mobile patrol were attacked by unknown personson 19.12.88 and 3 police constables and the driver were injured".The incident is admitted. However, it had no adverse effect on theconduct of the poll because the incident took place on the returnjourney after the ballot boxes had been taken to the polling station.
Paragraph (xvl) (g) "In the Wellawaya polling division at Buttala,police mobile patrol party was attacked by unknown persons andpolice sergeant 5060 and a reserve police constable were injured on19.12.88”. This incident is admitted. The evidence establishes thatthe Senior Presiding Officer and the election staff refused to proceedany further. Neither the ballot box nor the election staff reached thepolling station.
Paragraph (xvl) (h) "In the Moneragala polling division
Notices were put up at Moneragala warning people not tovote.
an unofficial curfew was enforced from 17th December1988.
bombs were exploded and guns were fired in theMoneragala town on 19.12.88 morning”.
The averments in (i) and (ii) are admitted. As regards (iii) theevidence does not show that the explosion of the bombs and thefiring took place in the Moneragala town.
Paragraph (xvl) (I) “In the Wellawaya polling division, the SLFPdivisional agent was forcibly prevented from entering the pollingstation at Weliyara”. The incident is admitted. There is no evidence toshow that this incident adversely affected voters or the conduct ofthe poll.
Paragraph (xvl) (j) “In the Wellawaya polling division on 19.12.88at about 1.30 p.m. P.C. 24217 Abeyratne Banda was on duty at the
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva. C.J.)
51
polling booth at Yudagamsuwa Junior School, when about 25 armedunknown persons in uniform had entered the polling booth and hadrobbed the guns of P.C. 24217 and R. P. C. Premeratne having tiedthe police officers and dashed the ballot box on the floor". Theincident is admitted. It is referred to in the petition in respect of thecharge of non-compliance".
The evidence adduced by the 1st respondent shows that as manyas 35 office-bearers and supporters of the UNP were killed during themonths of September, October, November and December 1988, inthe Moneragala electoral district. It is in evidence that families of well-known UNP supporters were done to death. D. M. Piyadasa,President of the UNP branch at Weliyaya was killed on 29.10.88. Aposter was found near his body stating “death for those who stoogefor the UNP". M. M. Loku Bandara, President of the Hulandawa UNPbranch was killed on 31.10.88. A poster lying by his body stated“death for those who stooge for the UNP. H. M. Sirisena of theWellawaya polling division who had served as a UNP polling agentsince 1965 was killed on 14.10.88. At the scene of the killing a posterappeared stating “punishment is death for those who support thetreacherous UNP". K. D. Keerthiratne, Committee Member of the UNPYouth League at Badalkumbura was shot and killed on 01.12.88. Hehad earlier received threatening letters asking him to stop working forthe UNP, but he had not complied with those orders. A posterappeared later saying that he was punished for working for the UNP.
Electoral District No. 22 – Kegalle
Paragraph (xvii) (a) “In the polling division of Yatiyantota SLFPbranch office at Ambanwela was set on fire on 18.12.88". The witnesswho testified to this incident is Kodikara, the SLFP organizer forYatiyantota. He. stated that he made a complaint to the police but atthe inquiry before the A.S.P. he withdrew the complaint. We are of theview that the evidence on record is not sufficient to prove the incident.
Paragraph (xvii) (b) “in the polling division of Rambukkanaunknown persons threatened voters and set up explosives aroundthe polling booth at Parape (No. 26) on 19.12.88". It is admitted thatsounds of explosions were heard around the polling station, but there
52
Sri Lanka Law Reports
[1992] 2 Sri LR.
is no acceptable evidence to prove the allegation that voters werethreatened.
Paragraph (xvfi) (c) "In the polling divisions of Yatiyantota -Deraniyagala on 18.12.88 the SLFP branch offices at Teligama andKitulgala were destroyed. In Yatiyantota polling booth at SiriwardenaBalika Vidyalaya (No. 01) SLFP polling agents were not allowed toattend to their duties". It is admitted that the SLFP branch office atTeligama was destroyed. The evidence of witness Kodikara issufficient to prove the attack on the SLFP branch office at Kitulgala.There is no satisfactory evidence to prove the allegation that theSLFP polling agents were not allowed to attend to their duties at theSiriwardena Balika Vidyalaya (No. 01).
Paragraph (xvii) (d) It is alleged that in the polling division ofAranayaka, (i) on 05.12.88 the SLFP electoral office at 369, DippitiyaBazaar at Aranayake was set on fire, (ii) on 08.12.88 the stageconstructed for the SLFP mass meeting was set on fire andM. Dayananda of Podapa who was guarding the stage was shot andkilled, (iii) the polling station at Wakirigala Raja Maha Viharaya(No.04) was damaged by bombs being thrown at it on 19.12.88before polling started. All three incidents set out above are admitted.
The evidence discloses that there was a fair amount of intimidationdirected against the SLFP in the Kegalla electoral district. Noevidence was adduced by the 1st respondent in respect of theKegalle electoral district.
We have outlined above the facts and circumstances relied uponby the petitioner and the 1st respondent in relation to the charge ofgeneral intimidation which is the charge that was pressed before usby Mr. de Silva in his closing address. However, it is right to statehere that neither Mr. de Silva nor Mr. R. K. W. Gunesekera made anysubmissions on the several acts or instances of violence and threatswhich the 1st respondent claimed were directed at his party and hissupporters. Indeed the cross-examination of the witnesses called onbehalf of the 1st respondent shows that by and large the incidents assuch were not denied. What was suggested was that those incidentshad no connection whatever with the Presidential election; the
sc
Sirimavo Bandaranaike v. Ranaslnghe Premadasa and Another
(G. P. S. de Silva, C.J.)
53
killings, attacks and threats were not politically motivated. Thesuggestions however were flatly denied by the witnesses themselves.They remained as mere suggestions wholly unsupported byevidence.
On a careful consideration of the totality of the evidence placedbefore us relating to the charge of general intimidation, it appears tous that the thrust of the JVP violence was directed against the UNP.Between the period 17.09.88 and 19.12.88 (16.09.88 being the dateon which the Working Committee of the UNP chose the 1strespondent as the candidate, according to the evidence on record)as many as 413 organizers, office-bearers and supporters of the UNPwere killed, and 237 were attacked. For the same period 32 SLFPorganizers, office-bearers were killed and 23 of them were attacked.The acts of violence against the UNP were spread throughout §0polling divisions in 15 electoral districts, whereas the anti-SLFPincidents occurred in 23 polling divisions in 13 electoral districts.Further, the incidents against the UNP were spread over a longerperiod of time, having regard to the evidence on record. Theevidence reveals that the numerous threats, killings and attacks onlocal party organizers and office-bearers of the UNP branches at thevillage level resulted in a serious and irreparable setback to theorganisation and the campaign of the 1st respondent. In additionthere was considerable evidence of resignations from UNP branchesby office-bearers and even ordinary members.
These resignations were consequent upon threats conveyed byletters. Several of these threatening letters were marked in evidence;letters which called upon the people not to work for and support theUNP – vide 1R 41. The document 1R 42 refers to the UNP as a“banned party”, “orders” office-bearers and members to resign from“this traitorous organisation”, and upon failure to do so “sentence ofdeath would be carried out”. 1R 43 is another significant documentwhich bears the heading “Banning of the United National Party". Itreads thus:- “The United National Party which has been traitorous tothe motherland is banned with immediate effect. All members arerequired to resign from membership and from all the offices theyhold. All persons should cease to lend any kind of support to the
54
Sri Lanka Law Reports
(1992] 2 Sri LR.
banned UNP. Death for those who violate the above conditions – JointCommanding Headquarters of the Patriotic Peoples’ Armed Troops”.These documents single out the UNP as the target of attack. Besides,there were threats directed at office-bearers and members of the JSSand large numbers were compelled to resign. It is in evidence thatthe JSS actively supported the UNP at previous elections. Many ofthose who were ordered to resign from the party or the JSS were alsodirected to put up “banners" and notices in public placesannouncing their resignations – vide 1R 95,1R 109, 1R 129,1R 138,1R 147, 1R 148, 1R 150,1R 151,1R 155, 1R 160, 1R 162, 1R 163, 1R164 and 1R 167. It is natural that all this would have had a strongadverse effect on supporters of the 1st respondent at the Presidentialelections. We are satisfied that the oral and documentary evidenceon record establishes that the weight of the JVP intimidation andviolence was directed at the UNP and its supporters and this hascontributed in no small measure to the low voter turn-out on 19.12.88.
There is another relevant matter to which we must refer. Mr. Choksydrew our attention to paragraph 05 of the petition wherein it isaverred that according to the results declared by the Commissionerof Elections, the majority by which the 1st respondent won is 279339•votes. It was the submission of Counsel that even if the petitioner gotone more vote than the majority obtained by the 1st respondent shecould still not have been declared elected. Mr. Choksy contendedthat the petitioner in order to win had to get the total votes receivedjointly by the UNP and the SLMP plus one more vote. Thus she wouldhave had to get 515059 more votes than she polled in order to havesucceeded at the election. It appears to us that this submission iswell-founded.
The question then is, upon a review of all the evidence, whetherthe acts or instances of intimidation had the requisite effect, namely,that the "majority of electors were or many have been prevented fromelecting the candidate whom they preferred”. In the preliminary orderthis Court has already ruled on the “true meaning" of these words ins.91 (a). The burden of proof is clearly on the petitioner in terms ofs.91 (a). However slight that burden may be, (having regard to the
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G.P.S.de Silva. C.J.)
55
use of the words “were or may have been prevented") yet the burdenof proof remains on the petitioner. We do not agree with Mr. de Silva’scontention that the word "may" also envisages the existence of “maynot” and is not inconsistent with it. The petitioner cannot leave thisimportant ingredient of the charge in doubt and yet claim that theburden has been discharged. Considering all the evidence in thecase, we hold that the petitioner has not succeeded in establishingthat the “result of the election may have been affected (1989 1 Sri
R. 240 at 270)(,). Accordingly, the charge of “general intimidation"relied on by the petitioner as a ground of avoidance of the electionfails.
We wish to make it clear that in arriving at the above conclusionwe have not taken into consideration the results of the Parliamentaryelection held in February 1989, although this was an item of evidencevery strongly relied on by Mr. Choksy. It was the contention ofMr. Choksy (i) that the evidence shows that there was less violence atthe Parliamentary election in February 1989 than at the Presidentialelection of 1988, (ii) consequently there was a larger voter turn-out atthe Parliamentary election, (iii) a comparison of the results of theDecember 1988 Presidential election and the Parliamentary electionof February 1989 shows that the bulk of the “extra votes” cast inFebruary 1989 were in favour of the UNP; this proves that it was theUNP that stood to gain when there was a decline in the violence.However, as pointed out by Mr. R. K. W. Gunasekera, the evidence toestablish that there was less violence in February 1989 is of atenuous nature. At the Parliamentary election there were a fairnumber of parties and a large.number of candidates. Some of theparties and "independent groups" did not “field” candidates on anislandwide basis. And more importantly, this was an election heldsubsequent to the Presidential election where the 1st respondenthad already been declared elected. Having regard to all thecircumstances, we agree with the submission of Mr. Gunasekera thatit is quite unsafe to draw arty conclusions from the results of theParliamentary election of February 1989, notwithstanding its proximityin time to the Presidential election.
56
Sri Lanka Law Reports
[1992] 2 Sri LR.
The charge relating to “other circumstances” as a ground ofavoidance of the election under Section 91 (a) of the Act:*
We now turn to the only other ground of avoidance relied on bythe petitioner. It is founded on s.91 (a) and is set out in the petition inthe following terms:-
“that by reason of other circumstances to wit, the failure ofthe Commissioner of Elections (the second respondent)and/or certain members of his staff to conduct a fair and freeelection, in accordance with the provisions of the PresidentialElection Act No. 15 of 1981, more particularly set out inparagraph 9 read with paragraph 8 hereof, the majority of thesaid electors were or may have been prevented from electingthe candidate whom they preferred”.
At the preliminary hearing the respondents filed objection to thisground of avoidance and contended that matters relied upon asconsisting non-compliance with the provisions of the PresidentialElection Act and as grounds for avoiding the election under s. 91 (b)cannot, as a matter of law, be included as a ground for avoiding theelection under s.91 (a). The Court considered this contention in thepreliminary order and overruled the objection. The Court reasonedthus:- “In paragraphs 9 and 10 of the petition the petitioner seeks torely on the instances enumerated under the head of non-compliancewith the provisions of the Election Law as “other circumstances" andpleads that by reason of their occurrence, the "majority of electorswere or may have been prevented from electing the candidate whomthey preferred”. In other words, the petitioner is also seeking to avoidthe election on the ground of avoidance set out in s.91 (a) of the Act,relying on non-compliance with the provisions of the Election Law…The words ‘other circumstances’ are wide enough to includeinstances of non-compliance with the law relating to the conduct ofelections. The petitioner was therefore entitled to plead instances ofnon-compliance to sustain a charge under s.91 (a) of the Act". (19891 Sri L.R. 240 at 281 and 282)(,). This ruling is very clear and isbinding on us. The trial proceeded on the basis of this ruling.
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva, C.J.)
57
Mr. de Silva, however, in his closing address deviated from thisruling and submitted that the “other circumstances” on which herelies consists not only of the instances of “non-compliance” but alsoof acts of general intimidation. In his written submissions counselstated:- "The 2nd ground on which avoidance of the election is alsosought to be founded on the same legal provisions, viz. s. 91 (a) butrests on a somewhat different factual basis. The factual basis here iscomposed of a combination of the acts of “general intimidation”referred to in paragraph 7 of the petition and the 'other facts andcircumstances' referred to in paragraph 9 read with paragraph 8. Thecontention is that when the acts of general intimidation are taken inconjunction with the evidence of the breakdown of electoralmachinery in various parts of the Island … they together constitutethe cause for the majority of voters being prevented from electing thecandidate of their choice. In a sense therefore there has been aninteraction of causes…”. This ground of avoidance was notformulated on the above basis in the petition and this clearly is notthe way in which it could have been understood by the respondentshaving regard to the ruling in the preliminary order. We are thereforeof the view that the petitioner cannot be permitted to present a case(at the stage of the closing address) which was not pleaded in thepetition and which is clearly contrary to the ruling given by this Court.The “other circumstances" must necessarily be confined to theinstances of "non-compliance" pleaded in the petition.
Mr. Marapana, counsel for the 2nd respondent submitted thatthere are 253 instances of “non-compliance" upon which evidencehas been led by the petitioner and these relate to 253 polling stationsin different parts of the Island; this includes the 49 polling stations inthe Moneragala electoral district where the poll was declared null andvoid by the 2nd respondent. There were altogether 8025 pollingstations in the Island at the Presidential election. Therefore there isno complaint in respect of 7772 polling stations. Evidence was led inrespect of only 3.1% of the total number of polling stations.
The complaints in respect of “non-compliance” may be verybroadly categorized as follows:-
(a) 96 polling stations which were either opened late or closedearly and some were opened late or closed early. It was the
58
Sri Lanka Law Reports
[1992] 2 Sri L.R.
submission of Mr. Marapana that these acts of “non-compliance” resulted in the loss of 288.75 polling hourswhich works out on an average to the loss of 2 minutes perpolling station island-wide. The evidence led on behalf of thepetitioner suggestive of the effect of this “late opening” or“early closure’ was the number of persons remaining in thequeue at the close of the poll (and thus unable to vote).Having regard to the number of such voters left in the queueat the close of the poll Mr. Marapana submitted that the totalnumber who were unable to vote was 4450 which constitutes.047% of the total number of registered voters. Nosubmissions to the contrary were made on behalf of thepetitioner.
There were allegations of inadequate staff in 38 pollingstations which constitute .47% of the total number of pollingstations. Mr. Marapana submitted that having regard to thenumber of persons left in the queue of voters at the close ofthe poll, 2915 voters were unable to cast their votes, i.e..031% of the total registered voters. Counsel for thepetitioner made no endeavour to challenge these figures.
In 49 polling stations in the Moneragaia electoral district nopoll was held at all and there were 44850 registered voters atthese polling stations. These 49 polling stations fall within thepolling divisions of Bibile (8 polling stations), Moneragaia (22polling stations) and Wellawaya (19 polling stations). Havingregard to the total number of registered voters at thesepolling stations and the average poll in each of the pollingdivisions it was the submission of Mr. Marapana that the totalnumber of voters “affected" is 8014. Again, there wereallegations of “non-compliance” in 5 polling stations in theMoneragaia polling division and 4 polling stations in theBibile polling division. Mr. Marapana submitted that thenumber of voters “affected" in those polling stations would be3210. Thus the total “affectation” in respect of voters (thoseunable to vote) in the Moneragaia electoral district is 11224.No submissions were made on behalf of the petitioner in thisregard.
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P.S.de Silva, C.J.)
59
There were 63 polling stations which were 'shifted', that is thelocation was altered. Section 4 (4) which permits such'shifting' reads thus:- “Where due to any emergency it isnecessary that the situation of any polling station should bedifferent from that specified in a notice published under sub-section (1), the Commissioner may cause the situation of thatpolling station to be altered in such manner as he may, in hisabsolute discretion, determine". While the “shifting" is notdenied by the 2nd respondent, the fact that an “emergency"had arisen was not seriously contested by the petitioner. Thereal dispute relates as to whether notice was given oradequate notice was given to the voters. In this regardMr. Marapana led evidence of “announcements” being madeby police officers and Army personnel on mobile duty andalso of notices put up at places notified in the Gazette interms of s.4(1). It was the submission of Mr. Marapana that ina fair number of these ‘shifted’ polling stations the votingfigures themselves indicate that the voters have been givenadequate notice of the change of location. In others, the pollwas very low but this may be due to the change of location orto other factors. The burden lies on the petitioner to establishthe nexus between the alteration of the location and the lowvoter turn-out. The evidence does not show that the petitionerhas discharged this burden. It was the submission ofMr. Marapana that the "affectation” of voters consequentupon the "shifting” of polling stations does not exceed 14495.No submissions to the contrary were made on behalf of thepetitioner.
Apart from the above instances of “non-compliance” there wereother "incidents” at 7 polling stations:- (i) The Senior Presiding Officerat the Ekala Maha Vidyalaya polling station (No.40) in the Ja-Elapolling division had torn off about 25 ballot papers from 25 differentbooks and placed them in a bag without issuing them (Paragraph viiat page 21 of the petition). There is no evidence as to what was donewith those 25 ballot papers after they were put into the bag. Thewitness who speaks to this incident says that he does not know “forwhose benefit the S.P.O. did this”. The maximum effect of thisincident is that 25 votes which should not have been included in thecount may have been included.
60
Sri Lanka Law Reports
[1992] 2 Sri L.R.
The Elamalpotha polling station No.18 in the Dambulla pollingdivision was attacked and consequently a large number of voterswere unable to vote. (Paragraph ix (a) at page 21 of the petition) Thissame incident has been pleaded under the “general intimidation"charge. There is no evidence to establish this incident. Weaccordingly hold that it is not proved.
At the polling station Ganhela (No. 16) in the Akuressa pollingdivision at about 11 a.m. a number of persons came in 3 jeeps andforcibly obtained from the Senior Presiding Officer 25 ballot bookscontaining 1250 ballot papers and the marked ballot papers were putinto the ballot box. (Paragraph x (b) at page 22 of the petition) Thisincident was not challenged. There is no evidence to show in whosefavour the ballot papers were marked. Nor is there evidence toindicate compliance with s. 35 (2) (c) of the Act. If there was no suchcompliance the probabilities are that the ballot papers would havebeen rejected by the counting officer – vide s. 51 (1) (a). It isdoubtful whether this incident would constitute an instance of non-compliance. Here again no submissions were made by counsel forthe petitioner. The evidence does not establish that the “result” mayhave been affected.
At the polling station at the Minhath Maha Vidyalaya, Dickwella(No.38) in the Devinuwara polling division around 12 noon about 10unauthorised persons entered the polling booth and forcibly obtained12 ballot books, containing 600 ballot papers, each of which wasthen marked with a cross and put into the ballot box. (Paragraph x (c)at page 22 of the petition) The evidence here was that the ballotpapers were marked in favour of the UNP. There is no record of thisincident in the journal P31. However, there is sufficient evidence toestablish this incident and we accordingly hold that the incident hasbeen proved. The “affectation” here would amount to 600 votes.
At the Bambarawewa polling station (No.16) in the Amparapolling division 50 ballot papers in excess of the number of voterswho came to cast their vote were found in the ballot boxes.(Paragraph xiv, page 24 of the petition) One witness was called totestify to this incident but his evidence does not show that there were
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva. CJ.)
61
50 ballot papers in excess. We are of the view that this incident hasnot been proved.
No ballot boxes reached the Dikwewa polling station in theKalawewa polling division. (Paragraph d (ii) at page 34 of thepetition) This allegation is not proved in view of the evidence ofDassanayake, the Government Agent, Anuradhapura, who statedthat on the morning of 19th December this polling station was shiftedto Maha llluppalama for security reasons.
At the S. Thomas’ College polling station (No.17) in theBandarawela polling division around 2.30 p.m. a number ofunauthorised persons forcibly entered the polling booth and forciblyobtained the petitioner’s polling agent’s list of voters and shouted atthe voters to vote for another party. (Paragraph xviii (b) at page 35 ofthe petition) Witness Somawathie the SLFP polling agent stated incross-examination that as a result of this incident no voters wereprevented from voting and that everyone present was able to vote. Byreason of the incident no adverse consequences were thusestablished.
We pass on to the more general allegation of “non-compliance" setout at paragraphs 8 (i), (iii), (iv) and (v) of the petition. (No evidencewas led in respect of paragraph (ii).
Paragraph 8 (i) avers that the 2nd respondent failed to comply withthe provisions of s. 21 (2) of the Act and in accordance therewithappoint another date for the taking of the poll in the electoral districtsof Matale, Matara, Hambantota, Kurunegala, Polonnaruwa, andMoneragala notwithstanding the outbreak of widespread violence formany days prior to the election and on election day. In consideringthis allegation, it has to be borne in mind that the Constitution and thePresidential Elections Act stipulate a period of time within which theelection has to be held. In accordance with the provisions of Article.31 (3) of the Constitution the last date for the poll would have been3rd January 1989. Thus the 2nd respondent could not havepostponed the holding of the poll in any electoral district beyond
that is, for a period of not more than 15 days. See also s.21
of the Act. What is of relevance for present purposes is that there
62
Sri Lanka Law Reports
[1992] 2 Sri L.R.
is no evidence to show that the climate of terror alleged in theseelectoral districts would have declined and the situation would haveimproved between 19.12.88 and 03.01.89. Nor is there any evidenceto indicate that the 2nd respondent had reasonable grounds tobelieve that the violence would be less within a period of 15 days. Inview of the time limit within which the poll had to be held, thepetitioner must show that the 2nd respondent had reason to believethat had he postponed the poll for any date before 04.01.89 therewas a reasonable prospect of the situation improving. We find nosuch evidence on record. Nor is there evidence that an applicationwas made for the postponement of the poll. On the other hand,Mr. Marapana submitted that the voter turn-out in the electoral •districts of Matara and Hambantota decreased by February 1989,indicating an escalation of violence. We accordingly hold that there isno basis for the alleged “non-compliance".
Paragraph 8 (iii) avers that the 2nd respondent declared null andvoid the polling in 49 polling stations in the Moneragala electoraldistrict without naming them and that he failed to comply with s.46Aof the Act. The fact that the poll was cancelled in 49 polling stationsis not contested. Mr. Marapana contests the allegation that the 2ndrespondent failed to comply with s.46A of the Act. The document 2R27 marked without objection at the trial clearly establishes that the2nd respondent consulted the election agent of the petitioner asprovided for in s.46A(8) of the Act. Paragraph 1 of 2R 27 states, interalia, that “This meeting is specially convened in conformity with s.46Aof the Elections (Special Provisions) Act No. 35 of 1988". Paragraph 8of 2R 27 reads thus:- “Both Mr. Ranjan Wijeratne and Dr. MackieRatwatte agreed with the views expressed by the Commissioner ofElections, that what would have polled, could not make a differenceto the result”. The election agent of the petitioner having thus agreed,it is not open to the petitioner to complain now. We accordingly holdthat there is no basis for this alleged instance of “non-compliance.”
Paragraph 8(iv) deals with postal votes. The allegation is that the2nd respondent failed to act in accordance with the provisions ofs.23 and in consequence “a number of persons who had the right tovote by post were unable to vote at this election, and the votes of alarge number of persons who voted by post were not counted at the
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G.P.S.de Silva. C.J.)
63
counting centres”. No evidence was led in respect of “postal voting"in the Colombo, Gampaha, Nuwara Eliya and Ratnapura electoraldistricts. The evidence adduced in the Kandy, Matale, Galle, Matara,Hambantota, Wanni, Batticaloa, Digamadulla, Puttalam,Anuradhapura, Polonnaruwa, Badulla, Moneragala and Kegalleelectoral districts does not prove the allegation contained in thepetition. The evidence led in regard to the Kurunegala and BadullaElectoral districts shows that no postal ballot papers were issued atall. The evidence does not indicate how the “non-compliance" mayhave affected the result of the election. No submissions whateverwere made by counsel for the petitioner in respect of “postal voting”.We accordingly hold that the allegation as pleaded has not beenproved.
Paragraph 8(v) refers to the failure to ensure that official poll cardswere sent to all registered voters as required by s.24 of the Act..Theallegation is that “as a result a large number of voters were preventedfrom voting”. The evidence is that in many polling divisions poll cardscould not be issued owing to the prevailing situation. It is also inevidence that the absence of a poll card does not mean that a voteris denied the right to vote. The Government Agent, Kandy, in hisevidence stated as follows:-
Q. It is not a requirement that a voter should be in possession of aballot (sic) card in order to cast his vote at the polling station?
A. That is so.
Q. Apart from the date of the poll the other particulars stated inthe poll card are also available in the voters’ electoral lists?
A. Yes.
Q. And these electoral lists are available to the candidates andtheir agents?
A. Yes.
Q. It is the practice as far as possible or convenient to continueusing the same location for a polling station…?
64
Sri Lanka Law Reports
[1992] 2 Sri L.R.
A. Yes.
Q. By which (sic) the voters know their polling stations quiteindependently of receiving polling cards?
A. It is generally known.
Q. Are you aware that the Commissioner of Elections causedradio announcements to be made of the fact that a voter wasentitled to vote although he was not in possession of a pollingcard?
A. Yes.
Q. The message was announced or telecast on the Rupavahanialso?
A. Yes.
There is no evidence to show that any voter was unable to cast hisvote because he had no poll card. Nor is there evidence to indicatethat a voter did not know the situation of the polling station becausehe did not get a poll card. In short, there is no evidence to suggestthat the result may have been affected on account of the failure toissue poll cards. The allegation as pleaded has not been established.
We have set out above the main instances of “non-compliance”relied on by the petitioner as “other circumstances” in terms ofs.91(a) of the Act. We have already held, in accordance with theruling of this Court in the preliminary order, that the burden is on thepetitioner to prove that by reason of the “other circumstances” theresult of the election may have been affected. This, the petitioner hasfailed to do; the evidence falls short of the required proof. In the resultthe second ground of avoidance relied on also fails. The petition isaccordingly dismissed with costs.
In terms of s.98 of the Presidential Elections Act No.15 of 1981, wedetermine that the 1st respondent was duly elected.
This is perhaps the longest trial held in this country. It continued forno less than 3 years. In this, not altogether easy case, we receivedthe full and complete assistance from all the counsel appearing for
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(G. P. S. de Silva. C.J.)
65
the petitioner and the two respondents. We wish to place on recordour deep appreciation of the comprehensive and cogent writtensubmissions on the law given by Mr. H. L. de Silva and the carefullyprepared, well-documented, and meticulously-arranged summary ofthe evidence handed over by Mr. Choksy as well as Mr. Marapana.These, considerably lightened the burden that lay on us.
RAMANATHAN, J. – I agree.
P. R. R PERERA, J. – I agree.
WIJETUNGA, J. – I agree.
S.B. GOONEWARDENE, J.
The Constitution of the Democratic Socialist Republic of Sri Lankaprovides for the office of The President of the Republic who isdescribed in Article 30 (1) as the Head of State, the Head of theExecutive and of the Government and the Commander-in-Chief of theArmed Forces. As the term of office of the then President was due toexpire on the 4th day of February 1989, the poll for the election of aPresident had, as required by Article 31(8) of the Constitution, to beconducted not less than one month and not more than two monthsbefore that date. The poll was fixed for and the election conductedon the 19th day of December 1988, a date which fell within the limitsprescribed, and the three candidates who contested were,Mrs. Sirimavo Bandaranaike of the Sri Lanka Freedom Party thepetitioner, Mr. Ranasinghe Premadasa of the United National Partythe 1st Respondent, and Mr. Oswin Abeygunasekera of the Sri LankaMahajana Party, no party in these proceedings.
The electorate for the purposes of the election was the wholecountry and, as required by section 3 (2) of the Presidential ElectionsAct No. 15 of 1981 (which is the principal statute governing theholding of an election of the President), divided into electoral .districtsand further subdivided into polling divisions and polling districts. Aswe were made to understand, the area of each such polling districtbroadly corresponded to the area served by a polling booth or to useanother term, a polling station. There were 22 electoral districts, 159polling divisions and 8025 polling districts.
66
Sri Lanka Law Reports
[1992] 2 Sri L.R.
The result of the election was declared by the 2nd Respondent theCommissioner of Elections in terms of section 56(2) of thePresidential Elections Act, that the 1st respondent was the candidateelected to the office of President and such result was duly publishedin the Government Gazette (Extraordinary) bearing No. 537/3 dated21.12.1988 thus:-
Oswin Abeygunasekera2357194.63%
Sirimavo Bandaranaike228986044.95%
R.Premadasa256919950.43%
Valid Votes5094778
Rejected Votes91445
Total Polled5186223
Majority279339
Total Registered Votes9375742
Total polled/Registered Votes55.32%
On the 9th day of January 1989, the petitioner Mrs. SirimavoBandaranaike, an unsuccessful candidate at this election, filed thispetition seeking a declaration by this Court that the election of the 1strespondent is void and/or undue. The broad grounds on which shehas sought this relief, and particularised in detail thereafter in herpetition in paragraphs 7,8 and 9 respectively, are contained inparagraphs 6A, 6B and 6C. They read thus:-
6 (A) That by reason of the occurrence of the incidents,hereinafter mentioned and the commission of the actshereinafter specified in paragraph 7 hereof, there was generalintimidation of the electors at the aforesaid election inconsequence of which the majority of the said electors were ormay have been prevented from electing the candidate whomthey preferred.
6 (B) That by reason of non-compliance with the provisions ofthe Presidential Elections Act No. 15 of 1981 (as amended) theaforesaid election was not conducted in accordance with theprinciples laid down in the said provisions and as hereinafterspecified and as particularised in paragraph 8 hereof, whichacts of non-compliance affected the result of the election andthe said election is in consequence null and void.
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
67
6 (C) That by reason of other circumstances to wit, the failure ofthe Commissioner of Elections (the 2nd respondent) and/orcertain members of his staff to conduct a fair and free election,in accordance with the provisions of the Presidential ElectionsAct No. 15 of 1981 more particularly set out in paragraph 9 readwith section 8 hereof, the majority of the said electors were ormay have been prevented from electing the candidate whomthey preferred.
Section 91 of the Presidential Elections Act, inter alia in sub-sections (a) and (b) provides that, on an election petition, the electionof a candidate to the office of President shall be declared to be voidby the Supreme Court, if the following grounds are proved to Court'ssatisfaction
that by reason of general bribery, general treating or generalintimidation or other misconduct, or other circumstances,whether similar to those before enumerated or not, the majorityof electors were or may have been prevented from electing thecandidate whom they preferred.
non-compliance with the provisions of this Act relating toelections if it appears that the election was not conducted inaccordance with the principles laid down in such provisions andthat such non-compliance affected the result of the election.
The allegations contained in any of the paragraphs 6A, 6B or 6Cof the petition, if successfully established according to law, wouldaffect the 1st respondent and no doubt therefore they are all ofconcern to him, but the case he was called upon to meet upon thepetition is contained in the said paragraph 6A.
The case the 2nd respondent the Commissioner of Elections wascalled upon to meet upon the petition however is contained in theother two paragraphs 6B and 6C of the petition.
The petitioner’s case in this petition as contained in paragraph 6Aof her petition, being that founded upon general intimidation asparticularised in paragraph 7 of such petition, falls to be decided by
68
Sri Lanka Law Reports
[1992] 2 Sri LR.
reference to the provisions of section 91(a) of the PresidentialElections Act.
The petitioner’s case as contained in paragraph 6B of her petitionis based upon the provisions of section 91(b), of the PresidentialElections Act and relates to complaints of non-compliance with theprovisions of the elections law as particularised in paragraph 8 of herpetition, the general heading of which reads “Non-Compliance withProvisions of Elections Law”.
The petitioner's case as contained in paragraph 6 C of her petitionis that the cumulative effect of all or a substantial number of theinstances and transactions enumerated in paragraph 8 was toprevent, in the manner set out in paragraph 9 of the petition, a freeand fair election being held and that by reason of their occurrence,there was another “circumstance" whereby the majority of electorswere or may have been prevented from electing the candidate whomthey preferred, thus constituting a ground of avoidance of theelection also under the provisions of section 91(a) of the PresidentialElections Act.
Before the evidence at the trial commenced, certain preliminaryobjections as to the maintainability of the petition as constituted,founded upon a claim of inadequacy of pleadings, had been raisedon behalf of both respondents, and in holding that the petition wasmaintainable in the form constituted, this Court examined theprovisions of sections 91(a) and (b) of the said Act. The order withrespect to those objections is to be found reported sub.nom.Bandaranaike v. Premadasa in (1989) 1 Sri Lanka Law Reports page240. I will at this stage proceed to mention in outline, the questionsaround which these objections had been examined, in particular thesubmissions of Mr. H. L. de Silva, learned Counsel appearing for thepetitioner, so that the general nature of such objections as well,insofar as that is necessary to be done, will also become apparent.
As regards the allegation of general intimidation relied on by thepetitioner as contained in paragraph 6A of her petition, the contentionon her behalf, as far as I can gather from the order on the preliminaryobjections, had been that the inclusion of section 91(a) in the
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
69
Presidential Elections Act was the way in which the concept of theEnglish common law, of a free and fair election had been introducedinto the statute. The argument put forward appears to have taken theform (though learned Counsel for the petitioner did submit thatcertain important aspects of his submissions are not reflected in theCourt’s order) that if proof could be adduced of general intimidation,which by reason of its magnitude deprived the electors of a free andfair election, it is possible to say then, that the majority of electorsmay have been prevented from electing the candidate whom theypreferred, and consequently that the Court is not called upon to enterupon the independent exercise of determining whether there is proofbefore it that the majority of electors may have been prevented fromelecting the candidate whom they preferred (implying the success ofsome other candidate), and that this is a matter not capable of proofin any accurate sense. The argument adopted, as one can gatherfrom the Court’s order, appears to have been on these lines, that ifgeneral intimidation is established and as a consequence a largenumber of voters are shown to have refrained from voting, thenecessary inference is that the majority of electors may have beenprevented from electing the candidate whom they preferred andtherefore what the petitioner had to prove was the existence of thatdegree of general intimidation which made a substantial number ofvoters keep away from the polls, thus eliminating free choice.
In its order on the preliminary objections, the Court, in examiningthe English common law concept of a free and fair election in thecontext of general intimidation, posed to itself (at p. 259) a questionthus:- “The question arises whether s. 91(a) of the Act embodieswhat Mr. H. L. de Silva P.C. described as ‘the pure and unadulteratedEnglish Common Law prior to 1949’ or as Mr. Choksy submitted ‘thatin addition to general intimidation etc., something more has to beproved’ by the petitioner to have an election avoided under section91(a)”. The Court answered this question (at page 262) so far as Isee, in the following terms:- "The case of the petitioner based on theground of avoidance under section 91(a) falls to be determinedsolely by a consideration and application of the provisions containedin section 91(a)”. The Court also said (at page 261) “We agree withMr. Choksy that mere proof of several instances or acts of general
70
Sri Lanka Law Reports
[1992] 2 Sri L.R.
intimidation would not suffice to avoid an election. In addition thepetitioner has to prove that these several acts or instances had theresult or consequence that the majority of electors were or may havebeen prevented from electing the candidate whom they preferred."The Court also posed to itself (at page 264) the following question“What is the meaning of the expression 'the majority of electors wereor may have been prevented from electing the candidate whom theypreferred’ ”. It answered such question (at page 270) in this form: “Ina case of general intimidation, the question that arises is, from theproved acts of intimidation of electors, is it reasonable to supposethat the result of the election may have been affected? This it seemsto us to be the true meaning of the words, ‘the majority of electorsmay have been prevented from electing the candidate theypreferred' ”. The following words the Court then immediatelyproceeded to add “But it will be open to the returned candidate toshow that the gross intimidation could not possibly have affected theresult of the election”.
With respect to the case of the petitioner as directed against the2nd respondent on the ground of non-compliance with the provisionsof the elections law, the submission made on her behalf at the inquiryinto the preliminary objections had been that upon a reading ofsection 91(b), it is apparent that there is no burden cast on her toestablish that the result of the election had been affected and thattherefore proof of such affectation is not an essential ingredient of theground of avoidance contained in such section. The tenor of thiscontention had been that when section 91(b) refers to “the result ofthe election"’ such reference can only be to a valid election and thatwhen the magnitude and extent of the non-compliance is to such adegree as to render the election a sham, the result goes with it. Thekey to the interpretation of section 91(b), it had been argued, is to befound in section 115 of the Act which is a provision guaranteeingprotection to the returned candidate and therefore that bothprovisions had to be read together and interpreted in a mannerconsistent with each other. The resultant submission therefore hadbeen that the words in section 91(b) which read “if it appears that theelection was not conducted in accordance with the principles laiddown in such provisions and that such non-compliance affected theresult of the election," had to be read to mean. “ if it appears that the
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene. J.)
71
election was not conducted in accordance with the principles laiddown in such provisions or that such non-compliance affected theresult of the election”. The Court upon an examination of earlierlegislation on similar lines and a consideration of the relevantauthorities concluded that it was not permissible to substitute theword "or” for the word "and” in that manner, and held that proof ofaffectation of the result was indeed an essential ingredient of theground set out in section 91(b) of the Presidential Elections Act.
At the stage of the final addresses in the case, learned Counselfor the petitioner Mr. H. L. de Silva contended that, as regards thecase of the petitioner based upon section 91(a) where reliance wasplaced on general intimidation, while not challenging the Court'searlier order or seeking its review, certain portions of such orderrequired clarification and that he would endeavour to demonstratehow they should properly be understood and that if so done, it wouldsupport his position. On the other hand learned Counsel for the 1strespondent, Mr. Choksy objected to any endeavour to reagitate thequestions decided upon by the Court’s earlier order, particularly thatportion of it which, in his submission, said that there was cast uponthe petitioner the burden of proving that the majority of electors wereor may have been prevented from electing the candidate whom theypreferred, which in essence he claimed, suggested an affectation ofthe result of the election, meaning a return of the wrong candidate. Inpoint of fact, as if in anticipation of such a possibility, Mr. Choksy, atthe stage at which he was addressing the Court, submitted that sucha course should not be permitted.
With respect to the ground of avoidance based upon section 91(b)of the Presidential Elections Act and founded upon a claim of non-compliance with the elections law, Mr. H. L. de Silva's position wasthat he was “abandoning his case". He submitted that he wasadopting that course by reason of the Court’s decision with respectto the preliminary objections, that when this ground of avoidance isinvoked, it must be proved affirmatively that the result of the electionwas affected, a virtually impossible burden as he suggested, and todischarge which there was no evidence in the case, as he impliedlyconceded.
72
Sri Lanka Law Reports
11992} 2 Sri LR.
Mr. H. L. de Silva adopted a new approach to the case of thepetitioner directed against the 2nd respondent, the Commissioner ofElections, on the basis of a failure to conduct a fair and free electionin accordance with the elections law as a ground of avoidance undersection 91(a) of the Presidential Elections Act, an approach new inthe sense that it was not the position taken up in the petition, as heindeed had to concede. He argued that non-compliance with theelections law was broadly, and subject to one exception (there washowever no clear statement as to what this exception was based on,nor any submissions made connected therewith), not something forwhich the 2nd respondent was accountable or responsible but thatthe acts of general intimidation complained of, resulted in abreakdown of the machinery of election on a large scale so that theinstances of such breakdown as pleaded, taken in conjunction withthe general intimidation which led to such breakdown, was another"circumstance", as that word is used in section 91(a) of thePresidential Elections Act and constituting a basis of avoidancethereunder. I will deal with this aspect of his submissions at anappropriate stage later on.
The contention of Mr. H. L. de Silva as to the question of generalintimidation relied on as a basis of avoidance of the election,standing by itself and without link to the pleaded items of non-compliance with the elections law is, as I see it, much the mostimportant aspect of the petitioner’s case. Whether as claimed, whatwas done by him can be described as an exercise calculated toachieve a true understanding of the Court’s order on the preliminaryobjections or whether, as was suggested by the reaction of opposingcounsel, it was in reality an attempt at reagitating some of the legalquestions upon which the Court had already ruled, I am of the viewthat the situation is one that can be dealt with, without embarkingupon a separate examination directed at resolving that initialquestion. The approach that commends itself to me as a satisfactoryone and one appropriate in the circumstances of this case is thus:The foundation upon which Mr. H. L. de Silva sought to rest the caseof the petitioner on this aspect is, that the expression in section 91(a),“the majority of electors were or may have been prevented fromelecting the candidate whom they preferred”, does not mean whatMr. Choksy contended to be its meaning, which is that the result of
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
73
the election was affected implying thereby the return of the wrongcandidate. I will hence first deal with the petitioner's case as if thesewords connote that which has been contended on her behalf tomean. Thereafter I will endeavour to ascertain what this expressionmeans, as I read and understand it, and in doing so attempt tointerpret what the Court intended these words to mean in its order onthe preliminary objections. Adopting such a course would have theadvantage of producing a final judgment arrived at, at the end of thecase, clarifying the issues upon the decisions of which the result ofthe case must be made to rest and which are seen at that concludingstage to arise with respect to all material propositions upon which theparties may be seen to be at variance, not forgetting that this is theCourt both of first and last instance, disposing of a matter which,notwithstanding its public interest dimension, has a good many of theattributes of a lis inter partes. Additionally I feel that thesesubmissions involve questions of a serious and important naturewhich deserve consideration.
Mr. H. L. de Silva, as I understood him, asserted what I will nowproceed to outline, but before doing so, in order to avoid confusionas to whether there is an intermix therewith of my own observations, Iwould make clear here that this paragraph will be confined entirely towhat he has said. He submitted thus:- The concept of a free and fairelection as known to the English common law before TheRepresentation of The People Act 1949 introduced certain significantchanges, is what is embodied in section 91 (a) of The PresidentialELections Act. The pivotal question (Mr. H. L. de Silva's words) istherefore as to the correct interpretation of section 91 (a). Theconcept embodies the right to choose freely and that presupposes achoice to be exercised without, duress, coercion or intimidation. Thecase of the petitioner is rested on a twofold basis of fact (as Mr. H. L.de Silva described it), but on a single legal ground. Therefore, toensure the petitioner’s success upon a proper discharge of theburden placed on her by section 91 (a), the following have to beestablished, namely, (a) the existence of a situation where asignificant number of voters were prevented from voting at theelection, which number should necessarily be numerically greaterthan the majority of votes secured by the 1st repondent, the returnedcandidate, which implies therefore that if this number is numerically
74
Sri Lanka Law Reports
[1992] 2 Sri L.R.
less, the election cannot be avoided, (b) that these persons who wereprevented from exercising their right to vote were so prevented,irrespective of the question of which candidate they would havevoted for, had they not been so prevented, and (c) that they were soprevented from voting by acts of general intimidation (that is undueinfluence, coercion, duress etc) of such a nature as would have beensufficient to deter persons of ordinary courage from voting and thatthey were thereby prevented from electing the candidate whom theypreferred. If these elements are established, then, irrespective of thequestion as to whom such acts of intimidation were directed at,irrespective of the question as to who the voters were, who were soprevented, that is whether they happened to be supporters of thereturned candidate or on the other hand supporters of either of theother two candidates, irrespective of the question for whom theywould have cast their votes, had they been free to vote, such proofconstitutes a sufficient basis for avoidance of the election undersection 91 (a) of the Act. Once these constituent elements areestablished the inevitable consequence is that the majority of theelectors are shown to have, as a matter of possibility, been preventedfrom electing the candidate whom they preferred, whoever thatcandidate might be. The other basis of avoidance relied on is thelarge-scale breakdown of the electoral machinery which, when takenin conjunction with the acts of intimidation established, togetherconstitute another cause for the majority of voters being preventedfrom electing the candidate of their choice, once again constituting aground of avoidance under section 91 (a) of the Act. It is reasonableto suppose that had it not been for these causes the voter turn-outwould have been around 80% of the total number of registered votersso that there was a Consequent shortfall of about two and a halfmillion votes. The evidence establishes that the 1st respondent had amajority of about 280,000 votes over those of the petitioner, andtaking account of the amounts of these two figures it is clearly seenthat those deprived of voting numbered six or seven times themajority secured by the 1st respondent over the petitioner, the runner-up. In that situation it is reasonable to conclude that the result “mayhave been different" in the sense that the majority of electors mayhave been prevented from electing the candidate of their choice,whoever that might have been. The point of divergence between therespective positions of the parties is as to whether there is, or there is
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
75
not, another requirement that it was incumbent upon the petitioner toestablish apart from the factors already mentioned, (which factors arethe acts of general intimidation committed together with thedemonstrated instances of the malfunctioning or breakdown of theelectoral machinery and the natural consequence flowing therefromwhich brought about the result that voters were unable to vote, andthe extent or magnitude of the resultant loss of votes being soconsiderable that the result may well have been affected had all thenon-voters cast their votes), yet another factor, which is that the resultof the election may have been affected, in the sense that but forthese acts another candidate other than the one declared returnedwould have been successful. Such an exercise to determine whethersuch an additional element exists would involve a computation ofnumbers which necessarily must be based upon conjecture andsurmise, being depended upon circumstance totally different andunpredictable, and therefore not something that the section couldreasonably be thought to.demand. The Court in any event would beprecluded from entering upon any such exercise by reason of theright to secrecy of voting which is ensured to all, not only by virtue ofthe relevant constitutional provisions and many provisions in thePresidential Elections Act, but also because any such exercise byviolating one of its essential components would erode the commonlaw concept of a free and fair election, which has found expression insection 91 (a) of the Presidential Elections Act. In thesecircumstances the petitioner's contention is that there is no burdencast upon her to show that, but for the acts complained of, either thatshe would have been the successful candidate or that someoneother than the 1st respondent would have been the successful one.
I have here in broad outline endeavoured to set out what I havebeen able to understand to be the position taken by learned Counselfor the petitioner.
Before considering what Mr. H. L. de Silva said as to the case thepetitioner has presented with respect to general intimidation, I find itconvenient to make reference to what he relied on as a finding in hisfavour contained in the order on the preliminary objections which heclaimed had the consequence that there was no requirement under
76
Sri Lanka Law Reports
[1992] 2 Sri L.R.
section 91 (a) of the Presidential Elections Act that calls upon thepetitioner to establish that the result of the election was affected, inthe sense of the success of another candidate other than the onedeclared elected.
Mr. H. L. de Silva’s submissions thereon were thus. Each of therespondents contended that the petition was defective, in that itcontained no plea as to what the petitoner's position was as regardsthe 45% of voters who did not vote at this election (55% havingactually voted), that is whether such contention was that if they had infact voted, they would have voted for her. The Court, he contended,held that there was no need to so plead, and since section 96 (c) ofthe Act demands that all material facts be pleaded, what is implicit inthat ruling is that it was not a material fact that had to be proved as tohow that 45%, if they could have voted, would in fact have voted.Since the Court, he argued, therefore considered that it was not amaterial fact, it could not be a requirement of the section, andtherefore it was not a part of the burden cast upon the petitioner toshow for which candidate the voters who did not vote would havevoted, had they the opportunity of doing so, and hence the section inno way demands that the petitioner has to establish that the result ofthe election was affected, in the sense that some other candidateother than the one declared elected would have been successful.This process of reasoning which learned Counsel adopted issomething, I find I am unable to go along with. The Court said (atpage 268) “In our opinion, how the majority were or may have beenprevented from electing the candidate of their choice need not bespecially pleaded”. The inclusion here of the word “specially” musthave the usual significance it would convey when used in a contextsuch as this and would suggest no sense different from “expressly"(as opposed to "impliedly"). Other passages in the order militateagainst the view that this passage was intended to do more thanstate that as a matter of pleadings, there need be no expressaverment of this. Examples of such passages are thus:- "(Therequirement that the petition) shall contain a concise statement of thematerial facts on which the petitioner relies” was "intended to securethat a respondent will know from the petition itself what facts thepetitioner proposes to prove in order to avoid the election and willthus have a proper opportunity to prepare for the trial… The termmaterial facts has a plain meaning in the context of requirements
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene. J.)
77
relating to pleadings, namely facts material to establish a party’scase", (at page 263 of the order on the preliminary objections); “Theobject of the requirement (as to what should be pleaded) is clearly toenable the opposite party to prepare his case for the trial so that hemay not be taken by surprise.” (p. 263). The passage then (at page268) that reads “The petitioner has, in her petition pleaded that thegeneral intimidation had this effect (that is that the majority were ormay have been prevented from electing the candidate whom theypreferred). In our opinion, how the majority were or may have beenprevented from electing the candidate of their choice, need not bespecially pleaded", must be understood, as I read these words, tolimit the question that the Court was concerned with at that stage, tothe adequacy or inadequacy of the pleadings.
As can be gathered from these passages, and in thecircumstances of this case I do not see it as a necessary conclusionderived from any process of legal reasoning to say that the petitionerwas not required to establish this requirement by reason only of theCourt having said that there was no need for a special plea in theway contended for the respondents, although generally as a matter ofpleadings, what has to be proved may well have to be pleaded. TheCourt exempted the petitioner only from pleading the requirement,specially, or expressly, as a prerequisite to proving it, and that too inthe circumstances of the case, having regard to what the petitionalready contained. When the Court ruled that the petitioner did nothave to aver how the majority were or may have been prevented fromelecting the candidate of their choice, it does not mean as claimed,that the Court thereby rejected a submission that there was anobligation to prove, either as a certainty or as a probability, how thevoters who were prevented from voting would have voted at thiselection had they the opportunity of doing so. I do not think that suchwas the conclusion which the Court reached as that contended byMr. H. L. de Silva and I therefore cannot, for myself, conclude thatwhat is stated merely as a matter pertaining to pleadings, can beextended in this way so as to say that what is implied is as to whatdoes not require to be proved. I therefore see no warrant for arrivingat a conclusion that this passage relied on by Mr. H. L. de Silva canbe read to mean that the Court ruled that there was no burden castupon the petitioner in that regard. Indeed the Court has not said soand I do not imagine that the Court would have intended to say so.
78
Sri Lanka Law Reports
[1992] 2 Sri L.R.
Furthermore, in the course of his submissions regarding what hetermed a clarification of certain aspects of the Court's order on thepreliminary objections, Mr. H. L. de Silva was heard to say somethingto the effect that the ratio decidendi of the order which is what is ofbinding force, had to be arrived at by reference to the questions inissue at that stage before the Court, in the sense of those upon whichthe Court had to rule, and that was as to the adequacy or inadequacyof the matters pleaded, that is whether in the state of the pleadingsas they were to be found, further proceedings upon the petitioncould or could not be continued. How therefore, it is possible toreconcile the statement that the ratio decidendi had to be construedin the way so argued, with this further submission that the Court heldthat there was no burden cast upon the petitioner as respects proofof this matter, which is that the majority of electors were or may havebeen prevented from electing the candidate they preferred asmeaning that the result of the election was affected, is something I donot clearly understand. Indeed having regard to the approach I haveadopted, it is not possible for me to draw the inference we werecalled upon to draw, there being not a single statement in the Court’sorder that there is no such legal requirement under section 91 (a) ofthe Presidential Elections Act.
It is convenient to get out of the way one other matter. In thecourse of his submissions Mr. H. L. de Silva emphasized that anintegral aspect of the right to a free and fair election is the right tovoting by secret ballot. Learned counsel used the word “precluded",with reference to any possible exercise the Court might indulge in,directed towards ascertaining, if I mistake not, whether any othercandidate would have been the successful one had the actscomplained of not been present, and in that context asserted that theprincipal obstruction to such a course would be the barrier erectedby the voters’ right to the secrecy of the ballot. It is therefore useful tosee whether the common law concept of a free and fair election asunderstood in England before The Representation of the People Act1949, demanded as an essential requisite or component thereof, theright to vote secretly. I do certainly understand that the constitutionaland other statutory provisions of our law, assure such a right tovoters. But the question my inquiry is directed to is as to whether thisright of voting by secret ballot, if taken away, would result in an
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
79
erosion of the common law concept of a free and fair election, which,of course, is the same as asking whether such a right is demandedas an integral part of that concept, which if I mistake not, is the effectof what Mr. H. L. de Silva claimed. Having regard to the submissionthat section 91 (a) is a statutory embodiment of the common lawconcept of a free and fair election, if the right to secrecy of voting isnot part to that common law concept, apart from such assurances ofa right to secrecy of voting granted by other provisions, it could notbe considered a requirement implicit in the section itself. The bestapproach I think to such a question is by reference to the judgmentof Lord Denning M. R. in the case of Morgan v. Simpson lU). Aquotation from that judgment (at pp. 725 and 726), other than as amatter of interest, bears repetition here because it helps tounderstand the background against which the early casesparticularly those decided before the Ballot Act 1872, some of whichI will refer to subsequently, had been decided. Lord Denning said:-
"The common law method of election was by show of hands.But if a poll was demanded the election was by poll: seeAnthony v. Seger (,5 A poll was taken in this way. Thereturning officer or his clerk had a book in which he kept arecord of the votes cast. Each voter went up to the clerk,gave his name and stated his qualification. The clerk wrotedown his name. The voter stated the candidate for whom hevoted. The poll clerk recorded his vote. (Sometimes the voterwent up with a card on which the particulars were written,and these were written down by the poll clerk). After the pollwas concluded, the votes were counted and the resultannounced. But the poll book was open to inspection. Then,if required, there was a scrutiny at which the vote could bechallenged, for example by showing a voter was not qualifiedto vote. In that event his vote was not counted. So the resultwas decided according to the number of votes cast whichwere valid votes. Sometimes the returning officer or his clerkmight refuse to record some of the votes without goodcause. If it were found that the rejected votes would havegiven a different result the election would be vitiated: seeFaulkner v. Eiger(,e). If they would not have affected the
80
Sri Lanka Law Reports
[1992] 2 Sri L.R.
result, the election was good, but the rejected voter couldhave an action for damages against the returning officer: seeAshby v. White (,7).
Such was the method of election at common law. It wasopen. Not by secret ballot. Being open, it was disgraced byabuses of several kind, especially at ParliamentaryElections. Bribery, corruption, treating, personation, wererampant. These were not investigated by the Courts of law.They were the subjects of petition to Parliament itself. Oftenmembers were unseated and elections declared invalid. Ifyou should wish to know what happened, you will find it inPower, Rodwell II & Drew’s reports of controverted elections(Power Rodwell & Drews Bection Cases 1846-1858 Vols. 1 &2) and in Charles Dicken’s Account of the Election onEatanswill (The Pickwick Papers ch. 13).
In 1868 the judges were brought in for the first time. Bythe Parliamentary Elections Act 1868 a petition to unseat amember was to be tried by a judge of one of the superiorCourts. He was to make a report to the Speaker: and hisreport had the same effect as that of an election committeepreviously. After that Act, the judges tried many election. petitions. Nearly all of them were for bribery and corruption,and treating. Most of them will be found in O'Malley &Hardcastle’s Reports (O’ Malley & Hardcastle Election cases1869-1934)…
Then in 1872 Parliament passed the Ballot Act 1872. Itrevolutionized the system of voting at elections. It providedfor voting by secret ballot".
A major objective of the change which was to introduce voting bysecret ballot had apparently been to eliminate the unsatisfactoryfeatures of the earlier common law system of voting by show, ofhands. I do not understand that the object of that change was tosupply a missing component in the common law concept of a free
sc
Slrimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
81
and fair election, so as to complete and perfect an otherwise perfectconcept, incomplete by reason of the absence of this right of votingby secret ballot. Indeed even the Ballot Act 1972 which for the firsttime introduced in England the system of voting by secret ballot, didnot ensure total secrecy to the degree that appeared to becontended for it of being the basis of the freedom of franchise. ThatAct itself contained a provision to enable a voter to be asked how hevoted, upon a scrutiny after a vote had been declared invalid. {Videthe observations of Grove J. in the Hackney Case 2 O' M & H 77 atpage 81).
Against a background of what I have said as to the positionregarding the method of voting under the common law which was aspointed out by show of hands, I will now make reference to some ofthe early cases which will show that the concept of a free and fairelection was much in the minds of the judges and recognised andgiven effect to as a concept full in all important respects, even whenthere was no secrecy of voting. A reproduction of excerpts from thejudgments in such cases will also help to understand the true and fullmeaning of the concept as the judges who decided these casesunderstood it, and here it must be pointed out that learned Counselfor the petitioner on many an occasion cited passages from suchcases as embodying the concept in all its dimensions.
The Drogheda Case (,8) was decided in 1869 before the BallotAct 1872 became law. The allegation in that case pertaining togeneral intimidation as contained in the petition was that its objectwas to secure the success of the candidate who was declaredelected. The contention of Counsel for the respondent had been thatif an organised and general system of intimidation was proved, stillbefore setting the election aside on such a ground, that it wasnecessary to establish that such intimidation, however excessive itmight have been, had a substantial influence on the fate of theelection. In elaboration of that argument it had been the submissionof Counsel that provided the respondent had an actual majority ofregistered electors, however small it was, then no matter whathappened outside, no matter how many voters were assaulted anddriven from the polling booth, no matter how many voters werecompelled to go by devious ways in order to get back to their homes.
82
Sri Lanka Law Reports
11992] 2 Sri LR.
no matter how much blood was shed, no matter how much spiritualintimidation had been brought to bear upon the electors, still if thecandidate who was returned could say that there were 1000 electorsin the Borough and no matter how, he had polled 501 of them, hiselection cannot be declared void on the ground of generalintimidation, although the unsuccessful candidate may, upon ascrutiny, by striking off individual votes on this ground show that, butfor the general intimidation he would have had a majority. Mr. JusticeKeogh in his judgment responded to that argument thus:-
“I must say at once that the argument put forward by theRespondent is one from which I wholly and entirely dissent. It issubversive, in my mind, of the whole principle of freedom ofelection. It is said by the Counsel for the Respondents, thatfreedom of election is secured provided the majority are shownto have had the power of recording their votes, I deny thataltogether. This was not solely a contest between theRespondent and the Petitioner. There is another and greaterinterest than belongs to either of them; there is the publicinterest. The humblest individual in the whole of theconstituency has as good a right without fear or intimidation tocome into the Court-house Upon the day of the election as therichest man upon the register, and as good a right as the greatmajority of the constituency. Take it that a candidate has by themost legitimate means obtained the votes of nine-tenths of theconstituency in his favour, yet it is of vital importance to thepublic weal that the remaining tenth should be able to recordtheir voles and to express their opinions. If the majority are notonly to send their own representative to Parliament, as of coursethe majority must do, but if they are to drive by terror and withignominy and with scorn and with denunciation the minorityfrom the poll, what becomes of freedom to this country?…
But I take it to be well settled law … that an organised systemeither of bribery or of treating will invalidate an election.
… to put general intimidation upon a parallel with generalbribery or general treating, it must be shown to spread oversuch an extent of ground, and to permeate through thecommunity to such an extent that the tribunal considering the
sc
Sirimavo Bandaranalke v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
83
case is satisfied, if it be so, that freedom of election has ceasedto exist in consequence. If that be the case, I for my part see nodistinction between an organised system of bribery, anorganised system of treating, and an organised system ofintimidation."
It is well not to overlook here, firstly, that the method of voting bysecret ballot had not yet become law and secondly, that theallegation in the case was that general intimidation of a widespreadnature was calculated to benefit a particular candidate.
The Bradford Case(19) decided in 1869, again emphasized that bythe common law, that is law not created by the enactments of Acts ofParliament, undue influence vitiates an election. Mr. Baron Martin saidin that case thus:-
“There are some influences which are called due influences,and other influences which are called undue influences, and thelaw has endeavoured to punish the use of undue influences.Amongst these influences there are what are called bribery,treating, and oppression, that is, an improper and unduepressure put upon a man. But if pressure is put upon a man, ora bribe is administered to him, no matter by whom, orrefreshments are given to a man, no matter by whom, for thepurpose of affecting his vote, the effect is to annihilate the man’svote, because he gives his vote upon an influence which thelaw says deprives him of free action; he becomes a manincompetent to give a vote, because he has not that freedom ofwill and of mind which the law contemplates he ought to havefor the purpose of voting. But that affects the man alone, it doesnot affect the candidate; it has merely the effect of extinguishingthe vote, and if there was a scrutiny for the purpose ofascertaining who had the majority of lawful votes, that man’svote ought to be struck off the poll, but that is all. But it hasbeen long held, before these Acts of Parliament passed at all,that by the common law of the land, that is, law not created bythe enactments of Acts of Parliament, bribery, undue influence,and undue pressure vitiate an election. So that if it had beenproved that there existed in this town generally, bribery to alarge extent, and that it came from unknown quarters, that no
84
Sri Lanka Law Reports
[1992] 2 Sri LR.
one could tell where it had come from, but that people werebribed generally and indiscriminately; or if it could be provedthat there was treating in all directions on purpose to influencevoters, that houses were thrown open where people could getdrink without paying for It, – by the common law such electionwould be void, because it would be carried on contrary to theprinciple of the law."
in the Salford Case m decided in 1869 Mr. Baron Martin said asfollows:-
“Before an election can be vitiated by reason of general riot andviolence, it must be shown to be such as to affect the freedomof election, which is that every person who has the franchiseought to be at liberty to go and have the means of going to thepoll and giving hjs vote without obstruction, and without fear orintimidation. To set aside an election on the ground of generalriot and violence, it must be established that personspossessing the ordinary nerve and courage of men have beenprevented from going to the poll to record their votes.’
The Stafford Case <2,) was decided in 1869, and it had beensubmitted by Counsel for the petitioner there, that sufficient had beenproved to show that the election ought to be declared void atcommon law on the ground of general intimidation. In response,Blackburn, J. said that he would admit that, if it was proved that therewas so much intimidation that the result of the election may havebeen affected, it was not necessary to prove that it actually wasaffected and that it is a.question of fact whether the intimidation hadbeen so great that it could fairly be said that it was not a free election-,that is, that if there had not been so much intimidation, such anumber of persons would have voted who did not vote, that the resultof the election would have been different.
In the Nottingham Case m decided in 1869 once again it wasemphasized that “if rioting takes place to such an extent that ordinarymen, having the ordinary nerve and courage of men, are therebyprevented from recording their votes, the election is void by theCommon Law, for the Common Law provides that an election shouldbe free in a sense that all persons shall have an opportunity of
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
85
coming to the poll and voting without fear or molestation. But for thepurpose it must be rioting to an extent certainly to deter a man ofJ ordinary reasonable nerve from going to the poll”.
The foregoing cases contained in the O'Malley & Hardcastle seriesof reports had all been decided before the Ballot Act of 1872 whenthe common law method of election by show of hand prevailed andthere was no voting by secret ballot. Secrecy of voting, as being akind of foundation upon which the concept of .a free and fair electionstood, if that was the contention, was not thought by the judges to bean essential concomitant of a free and fair election, as these casesshow.
The case of The Borough of Dudley™ decided after the Ballot Actof 1872 introduced voting by secret ballot, was one where thesubstantial allegation in the petition was that there was so much riotand intimidation by mobs that there was no free election. Mr. JusticeGrove set aside the election, in the view that he took on the evidence,that although the rioting and assaults were not committed by mobsacting for one side only, the more formidable violence was on the partof the mob which espoused the respondent’s side and that the resultwas that a large number of voters were deterred and prevented fromexercising the franchise and that the election therefore was not a freeone and the constituency had not a fair opportunity of freelyexercising the franchise. He went on to add however:
"I by no means say that if a case had been made out of theviolence being wholly or substantially on the side of thedefeated candidate, and if I was satisfied that the result of thepoll was a fair expression of the opinion of the constituency, Ishould have come to this conclusion, but the case, taking it inthe most favourable view for the Respondent, stops far short ofthis.”
In the North Durham case™ once again decided after voting bysecret ballot had been introduced into the law, the allegation in thepetition was that the election was void on account of general riot andintimidation. In declaring the election void, there being a concession
86
Sri Lanka Law Reports
[1992] 2 Sri L.R.
by the successful candidate that the evidence was adequate toestablish general intimidation, Mr. Baron Bramwell commented thus:
“I take it that the law is this; first of all, there is the statutoryintimidation, that contemplated by the statute, if one may usesuch an expression, that is, an intimidation contemplated by thestatute which avoids the seat, where a candidate or his agentis guilty of it. But besides that there is another intimidation thathas been called a common law intimidation, and it applies to acase where the intimidation ,is of such a character, so generaland extensive in its operation that it cannot be said that thepolling was a fair representation of the opinion of theconstituency. If the intimidation was local or partial, for instance,if in this case it had been limited to one district, … I have nodoubt that in that case it would have been wrong to have set.aside this election, because one could have seen todemonstration that the result could not possibly have beenbrought about by that intimidation, and that the result would nothave been different if it had not existed. I do not mean the resultof the polling in that particular district, but the general result ofthe majority for the Respondents. But where it is of such ageneral character that the result may have been affected, in myjudgment, it is no part of the duty of a judge to enter into a kindof scrutiny to see whether possibly, or probably even, or as amatter of conclusion upon the evidence, if that intimidation hadnot existed the result would have been different. What the judgehas to do in that case is to say that the burden of proof is castupon the constituency whose conduct is incriminated, andunless it can be shown that the gross amount of intimidationcould not possibly have affected the result of the election itought to be declared void. Now in questions of this sort onemust look not only to the amount of intimidation, but to theabsolute majority which has been obtained. It was the opinion ofMr. Justice Willes, and I believe it is not inconsistent with, theopinion of Mr. Justice Keogh, as expressed in that celebratedand most useful judgment which he gave in the Galway case,that you are to look at the probable effect of intimidation, whichconsists of two things, the extent and operation of theintimidation, and the majority which the sitting members got..
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
87
Now, I think if it were otherwise, and if one were told that partialintimidation would avoid an election, although it was certain thatit had not affected the result of the election, the consequencewould be that a few mischievous persons might upset everyelection."
The case of The Thornbury Division of The Country of Gloucester wonce again decided after the Ballot Act of 1872, was one where anallegation in the petition was inter alia of intimidation. Mr. Justice Fieldin delivering judgment said as follows:
“But besides this statutory prohibition there is what is knownas common law intimidation and riot. Now there are two leadingcases upon this subject, one the North Durham Case124’ and theother the Drogheda Casem and I propose here to adopt thelanguage of Lord Bramwell in the former, which confirms theview of Mr. Justice Willes in the Lichfield Case<26), as supplyingthe legal test which we must apply to these facts. Now in thisconstituency, out of twenty-three polling districts only three areaffected by this crime, and that out of 11333, the total number ofelectors, 9529 went to the poll. The number of voters in thethree districts in question is 789, and all but 87 voted. It istherefore difficult to come to the conclusion that any suchintimidation or violence was used as practically prevented anyconsiderable number of persons from voting. Again, we mustconsider not whether any particular person or particularlynervous person was affected by it. We must take the electors asan average of ordinary men who may be disinclined to go to thepoll, but who were not necessarily intimidated. A man ofordinary courage would not necessarily be intimidated by whathappened, and it id a very strong feature in the case that,speaking generally, the violence did not occur until after 6o'clock in the evening, and the petitioner has not attempted toprove that during the interval between 6 and 8 o’clock, personswho wanted to go to the poll were prevented from doing so, orthat practically there was no polling between these hours. I saythat if that had been proved it would have gone strongly to showthat this rioting had the effect of intimidating voters … It seemsto me that the question which I have to decide is whether all the
88
Sri Lanka Law Reports
[1992] 2 Sri LR.
electors of the other divisions of the constituency are to bedisfranchised for what was done in the three divisions, and afresh election held with all its turmoil and excitement. That willhave to be done if I am satisfied that there has not been thatfree exercise of the franchise which everybody is entitled tohave, and that the absence of that has been caused byintimidation and riot. But after the most careful consideration, Iam unable to come to the conclusion that this case falls withinthe principle of the Durham Case, and upon this part of thecase my judgment must be for the Respondent."
The North Louth Case® was heard in 1911. It came up before abench consisting of Gibson, J. and Madden, J. In the course of hisjudgment Gibson, J. observed that the election under considerationheld in the month of December 1910 was fought on the same registeras the one in the month of January previous, the total number ofregistered electors being 5761 at both such elections: that at theJanuary election the unsuccessful candidate in the Decemberelection won by a majority of 99, the total polled being 4786 leaving975 unpolied: that at the December election the poll was 4556leaving 1205 unpolled: that the decrease in number of actual electorswas 230, a figure which might represent natural wastage on theregister from lapse of time, and that the 'successful candidate at theDecember election was returned by a majority of 488. In hisjudgment Mr. Justice Gibson at pages 136 and 137 stated thus:-
“To upset an election for general intimidation it is necessary toshow that there was such general intimidation as might haveaffected the result of the election (the Stafford Case<21) theThornbury Case(27) The Ipswich Casem).
Where there is such general intimidation, the onus of provingthat the gross amount of intimidation could not have affected theelection is cast upon the constituency. The North DurhamCase'24’.
The amount of the majority is held to be important. Looking atthe figures in the various polling districts at the two elections on
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
89
the same register, a circumstance that has perhaps neveroccurred before, the restricted character of the intimidationproved as regards area and number of cases, the majority of488 on the same register as in January, I find it difficult to bringthe class of intimidation here relied upon, which I may termpreventive intimidation, keeping voters from the poll, within theauthorities as to general intimidation at common law.
Dealing with what is to me a new field of law, I am unwilling toextend the principle of common law intimidation by preventionbeyond what the authorities cited seem to cover…
The petitioners’ case was rested on preventive intimidation ofthe type dealt with in the reported cases relied upon by theRespondent's counsel. The difficulty in those cases was that somany of them depended on facts, and one case on facts couldnot bind another, as Lord Halsbury has so emphaticallyexplained in London Joint Stock Bank v. Simmons™ and LordChief Baron Palles in Rex v. Dolani30> and the difficulty isincreased when, as frequently happens the facts do not appearin the report but must be inferred from the judgments.
To avoid misconception I wish to point out that this point ofnumbers polled, districts, and majority, always difficult ofapplication, has, in my opinion, little, if any operation when theintimidation is not preventive, to exclude voters, but persuasive,to win votes by a process of forcible conversion addressed tofear.”
In the same case Mr. Justice Madden (at pages 172 & 173) saidthus:-
“Intimidation operates on the mind of the intimidated, andwhen this influence pervades the electors to such an extent asto render the action of the constituency other than free, theelection held under such circumstances is void and of no effectat common law, irrespective of any question of agency betweenthe authors of the intimidation and the candidate in whoseinterest it has been exercised. From its very nature it is
90
Sri Lanka Law Reports
11992] 2 Sri LR.
incapable of the exact proof which can be adduced with regardto particular instances of intimidation, or other undue influence.But its existence is a conclusion to be inferred from theevidence given in the case as well of general conduct as ofspecific acts, and from a consideration of the nature of theundue influence alleged to have been brought to bear upon theelectors and of the action of the constituency; a matter in regardto which the amount of the majority by which a seat was wonand the number of.electors polled, relatively to the entireconstituency and also to other elections, are proper to beconsidered”.
One of the objects of the exercise of examining these old Englishcases has been to ascertain the extent to which secrecy of votingwas considered an essential element of the English common lawconcept of a free and fair election, and taking into account what Igather from them and also keeping in mind the secrecy of votingprovisions contained in the Constitution and in the PresidentialElections Act, I find that I cannot read such provisions so as toimpose a prohibition upon the Court from examining any material thatmight be considered relevant, as to how or in what manner voterswould or might have voted, so long as there is not involved anexercise at asking a voter how he did in fact vote, although there too,I see nothing to prevent a voter volunteering that information.
I have excerpted such material from the judgments in the casesreported by O’Malley & Hardcastle from their compilation of decisionsin trials of Election Petitions, as I thought would be of use for a properunderstanding of the principles of the common law as to generalintimidation and their application to the facts of any given case. Someof them as pointed out were cases decided at a time when votingwas not by secret ballot. In some of them, if not in a good many ofthem, the general intimidation complained of appears to have beencalculated either to achieve the success of one candidate or thedefeat of another, a feature which perhaps could be distinguishedfrom the present case.
It is perhaps well to keep in mind that in the case we areconcerned with, the intimidation was not calculated to advance theinterest of the 1st respondent. For the sake of completeness therefore
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
91
it may be mentioned here that there is to be found the followingpassage in Rogers on Election 20th Edition Volume 2 at page 347which gives the effect of the authorities as to the position in Englandunder the common law thus:-
"Whatever the form taken by the violence, it would be asufficient answer to prove that it was practised not by personsacting in the interest of the respondent, but against him, and inthe interest of the other candidate..
Although in the instant case the intimidation was not practised inthe interest of the petitioner, yet such intimidation was not practisedby persons acting in the interest of the 1st respondent either, andtherefore to that extent shares this common feature with the situationcontemplated by this passage, the implication of which is somethingthat the Court should not be altogether unmindful of.
It is possible now to embark upon an examination of the essentialsubmissions relied on by Mr. H. L. de Silva as constituting the basisof the petitioner’s case. The most important aspect of thosesubmissions is on the question of general intimidation in respect ofwhich, in the context of section 91(a) of the Presidential Elections Actis to be found the following passage (at page 270) in the order of theCourt on the preliminary objections, which I have set out in partearlier but which for ease of understanding is reproduced here onceagain.
“So it seems to us that on the basis of instances or acts ofgeneral intimidation established by evidence, the Court maydraw a reasonable inference therefrom that the majority ofvoters may have been prevented from electing the candidate oftheir choice. In a case of general intimidation the question thatarises is – from the proved acts of intimidation of electors, is itreasonable to suppose that the result of the election may havebeen affected? This, it seems to us to be the true meaning ofthe words 'majority of electors may have been prevented fromelecting the candidate whom they preferred’. But, it will be opento the returned candidate to show that the gross intimidationcould not possibly have affected the result of the election”.
92
Sri Lanka Law Reports
[1992] 2 Sri LR.
The effect of what Mr. H. L. de Silva suggested with respect to thispassage was tantamount to saying that the inappropriate use of theword "result” there, has led to a good part of the confusion whichneeds clarification. He submitted that nowhere has the Court in itsorder equated the expression appearing in section 91(a) “the majorityof electors were or may have been prevented from electing thecandidate whom they preferred" with the words, “the result of theelection was affected", as used in section 91(b), such that the formerexpression must be taken to convey the same notion as the latter.Indeed his contention was that the Court could not, upon a correctunderstanding of the law have done so. The word “result" occurs intwo places in this passage and if I understood Mr. H. L. de Silvacorrectly, his complaint was that its use the first time is misleading.Once again, if I understood him correctly, the use of this word asecond time is appropriate and denotes that there is a burden caston the 1st respondent, the manner in which that burden was to bedischarged being what was indicated by the way the word “result”was used the second time. This latter part of his submission howeveris something I will reserve for consideration later. The use of the word"result” as appearing the first time in that passage was, hecontended, an “imprecise judicial paraphrase” of the true words insection 91(a). He submitted that the word “result” as used the firsttime in this passage by the Court, was intended to mean, “the effector consequence on the voters' freedom of choice of candidate" andthat the upshot of this at this election was of such a magnitude as tolead to the conclusion that the majority of electors may have beenprevented from electing the candidate of their choice, regardless ofwhom they would have voted for, if they had had the opportunity todo so. It will become necessary therefore at some stage, to examinewhether the Court's expression complained of, was intended toconvey what Mr. H. L. de Silva contended was the true meaning ofthe section or whether on the other hand it was truly intended toconvey the other meaning suggested by the "imprecise judicialparaphrase”. Mr. H. L. de Silva's position was that he was notcontending that it was no part of the petitioner's case upon thepetition to show that the majority of electors were or may have beenprevented from electing the candidate whom they preferred. Hisargument rather was that the expression did not mean the same as,“the result of the election was affected", as these words appear in
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene. J.)
93
section 91(b) of The Presidential Elections Act and as interpreted bythe Court in its order on the preliminary objections. These words, inhis submission, were intended to suggest the notion that a significantnumber of voters were prevented from exercising the franchise, thatis to say, that a greater number than those who voted wouldotherwise have voted freely but were prevented from doing so. Hecontended that when the Court in its order on the preliminaryobjections used words which might on their face suggest that theymean an affectation of the result, that was a judicial paraphrase ofconvenience and hence as I understood him, this inexactitude wasone of the matters in that order that require clarification. Theinterpretation sought to be placed on section 91 (a) of thePresidential Elections Act on behalf of the petitioner, Mr. H. L. de Silvasaid, calls for proof of the facts and circumstances of the generalintimidation and their intensity of virulence and perversive character,and the magnitude of the effect brought about, or, in other words thenumber of voters affected by being unable to vote, without the Courthaving to delve into their political loyalties or choice of candidate.The evidence while demonstrating the existence of generalintimidation of that quality and kind had the result, he claimed, ofpreventing about 25% of the total of registered voters numberingaround two and a half million from exercising their franchise, thatfigure being six or seven times the size of the majority which the 1strespondent secured over the petitioner. In these circumstances hecontended that there is by way of evidence in the case, sufficientmaterial in discharge of the burden which the section cast upon her.Something that strikes me here is, that the mere existence of asituation where the number of votes not cast is many times thedifference between the votes secured by the winning candidate andthe runner-up, should not, of itself, be allowed to assume an undulyimportant significance. It seems to me that the larger the electorate,as for instance when the whole Island is taken into reckoning as at aPresidential Election where therefore the total number of registeredvoters would run into millions, conceivably at any election (regardingwhich there is no complaint) the difference between the actual voterturn-out and a theoretically possible 100% voter turn-out could wellbe many time the difference in votes between those of the successfulcandidate and of the runner-up. At the same time it must not beoverlooked that the smaller the majority of the winning candidate over
94
Sri Lanka Law Reports
[1992] 2 Sri L.R.
the runner-up at any such election, the greater the chance there isof that number as a fraction of the shortfall in votes becomingwidened.
It is now necessary to consider what implications arise from Mr. deSilva’s claim that these factors have been established. For a startthere is the claim that reasonably, 80% of the total number ofregistered voters (i.e. the total of the 55% that voted and the 25%who were said to have been prevented from voting) could have beenexpected to have voted at this election, had it not been for suchgeneral intimidation. That percentage has been arrived at on thebasis that it represents what he termed the national average of voterswho have voted during the post-independence period.
In this connection before looking at what the evidence in the casetaken as a whole suggests as to general intimidation, there arehowever one or two prefatory matters that need to be mentioned.Section 91 (a) of the Presidential Elections Act requires that it mustbe demonstrated that by reason of general intimidation, the majorityof electors were or may have been prevented from electing thecandidate whom they preferred. What does the expression “by reasonof general intimidation” mean? The meaning as I have understood itsuggests, that as a consequence of general intimidation there wasthis effect, that the majority of elector's were or may have beenprevented from electing the candidate whom they preferred, which tomy mind connotes once again the notion that the general intimidationmust be calculated to bring about this effect. The meaning adoptedby the petitioner I find not inconsistent with this position and in herpetition she has chosen to use the words, “In consequence”. Mr. H.L. de Silva’s submissions also suggested this, that it was theobjective of this intimidation to bring about this effect. In the case ofPiyadasa v. Gunasinghe 43 N.L.R. 36 Hearne J. (at page 39) usedthe same word,"calculated" that I have used as the followingpassage shows “I have mentioned in bare outline the incidents priorto December 14 which were calculated to prevent and did, in myopinion prevent a free and fair exercise of the franchise."
General intimidation, as opposed to statutory intimidation canconsist of two broad types. Nagalingam, J. referred to them inTarnolis Appuhamy v. Wilmot Pereram in the following passage:
sc
Sirimavo Bandaranalke v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
95
"No evidence was given of what may be termed coerciveintimidation, that is to say, intimidation having for its object theuse of force or threat to compel a voter to vote for a particularcandidate, but what evidence was led was led to show that theelectorate was subjected to preventive intimidation, that is tosay, intimidation which had for its object the prevention ofelectors from going to the polls lest the rival candidate gets theirvotes”.
Nagalingam, J. refers to these two types of intimidation as being“coercive" and "preventive” in this passage somewhat on the linesthat Mr. Justice Gibson did in the North Louth case I have alreadyreferred to. In the context of the case before him, he has describedthe object of preventive intimidation to be to deter voters from votingfor the rival candidate, which suggests that such intimidation hadbeen practised for the benefit of his opponent. Apart from that kindof preventive intimidation, there can perhaps be another type, whichis general intimidation calculated to prevent all voters from going tothe poll, regardless of their voting preferences. It is that type ofgeneral intimidation that is of relevance in this case and therefore itbecomes necessary to see what the evidence suggests in thisregard. Mr. H. L. de Silva in the course of his submissions contendedthat there were admissions by Mr. Choksy to the effect that there wassuch general intimidation as had the effect he contended for.Irrespective of whether or not there be any such admission, it isessentially a part of the Court's function and indeed its duty to reachthe appropriate conclusions and draw the necessary inferencesarising from the evidence before it. In addition in a matter pertainingto an election petition, there is the public interest element involvedand the Court cannot by reason of admissions or anything to thateffect be totally relieved of its duty in this regard. (vide for examplethe approach the Court adopted in the North Louth Case(8). Even if itmeans that any important questions that may arise, not only asrespects general intimidation but also as to certain other aspects ofthe case have to be decided upon a basis that has not beencontended for by any of the parties, that is something that cannot behelped. In saying so I also have in mind what I consider to be thesomewhat unexpected turn the case took at the end. As a statementof general application, it is correct in any event to say that a Court is
96
Sri Lanka Law Reports
11992] 2 Sri LR.
called upon to arrive at a just decision in the case, and thatsometimes may have to be done regardless of what the parties mayor may not contend. Indeed a Court is fully empowered to act in sucha fashion as was pointed out by Sir John Donaldson M. R. in the caseof R v. Chief Constable of the Merseyside Police Ex parte Calveleyand others(31>.
“In reaching this conclusion, the Court, as it was entitled todo, was proceeding of its own motion in the sense that thiswas not a contention advanced on behalf of either party".
Mr. R. K. W. Gunasekera who continued the submissionscommenced by Mr. H. L. de Silva for the petitioner, made twostatements, one as his own assertion, and the other in response to aquestion put by me. His assertion was that the contention ofMr. Choksy that the situation during the 1989 general election hadundergone an improvement from the standpoint of generalintimidation, as compared with what it was at the time of thePresidential Election in December 1988, was incorrect. My questionto him that resulted in the other statement was as to what inference,in the state of the evidence before the Court, one could draw from theincrease in the percentage of those who voted in the 1989 generalelection, as compared with the percentage that voted at thePresidential Election in 1988 and his reply was that there was noinference which such evidence suggested. The effect of the twostatements together, to my mind, is to suggest that there was nochange in the atmosphere of intimidation at the different times atwhich these two elections were held, but that there was nonethelessan increase in the percentage of voter turn-out, a phenomenon whichaccording to what Mr. Gunasekera said, the evidence did not explain.One would not then be too far wrong in saying that if such could bethe position as to the increased percentage of voting in 1989 byreference to 1988, similar unknown and unascertained causes couldwell have also contributed in some measure to a decline in thepercentage of those who voted at the Presidential Election of 1988 ascompared with the so-called national average of 80%. I mostcertainly must not be understood to say here that there is noevidence to show that general intimidation did have an effect on thevoter turn-out in 1988. What I do say is that to state a percentage
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
97
such as 80 points on the basis of a national average, and then toattribute the whole of the claimed shortfall to general intimidation,could well be inaccurate. The petitioner’s own contention is that therewas this climate of intimidation which was prevalent before andduring the Presidential Election in 1988 and continued thereafter,even beyond the time of the general election in 1989. What is it thenthat the evidence suggests? If one takes account of some items ofevidence, it could be said that there were calculated attempts atintimidating voters in order to prevent them from voting. Examples ofsuch evidence were of posters displayed warning against voting atthe election, disruption of political meetings, attempts to sabotagethe holding of the poll at some polling stations, preventing the pollfrom being conducted at certain polling stations etc. But apart fromthese items of evidence which suggest the adoption of tactics ofgeneral intimidation calculated to prevent the holding of the election,the remaining mass of evidence demonstrates something somewhatdifferent. Mr. H. L. de Silva’s description of that at different placeswas thus:-
"An election held amidst a massive compaign of terror, heldunder an all-pervasive climate of fear”.
"All these acts have been commonly attributed to the JanathaVimukthi Peramuna which long before 1988 had given up anyhope of coming to power through the parliamentary processwhen the leader of the J.V.P. who was a candidate at the 1982Presidential Election had only succeeded in obtaining less than6% of the votes polled. The violence that began with the protestagainst the 1987 Indo-Sri Lanka Accord increased in intensityand continued throughout 1988. It is beyond dispute that the
J.V.P. was opposed to the holding of the Presidential Election in1988 not because it was opposed to any particular candidatebut because the entire objective was to bring about arevolutionary change in the legal order and overturn the entireconstitutional structure which provided the legal machinery forthe establishment of a governmental regime. The J.V.P.opposition and attacks were directed at the very holding of theelection because the objective was to destroy the entire systemof democratic government. It was an anti-systemic movement
98
Sri Lanka Law Reports
[1992] 2 Sri L.R.
calculated to destroy any claim to legitimacy through a processof democratic elections. Hence the carefully planned attemptsto disrupt the civil administration, paralyse the transport andcommunications systems, impose unauthorised curfews, thecompulsory closure of shops, business houses, schools,hospitals and bring about a general breakdown in essentialservices. In short, the object was to bring about a situation ofwholesale chaos and disorder that would have enabled arevolutionary take-over of the administration”.
“As I said before it was an anti-systemic agitation to dismantledemocracy and usher in a reign of chaos and confusion beforethe final push for a revolutionary take-over of the administration.I think all of us here can vividly recall the tense atmosphere thatprevailed at the time. We remember the anxiety, fear andapprehension which afflicted everyone and filled our nights anddays. The whole nation was in the grip of fear and traumatisedby the J.V.P. terror. A great many of those incidents have beenunfolded in the evidence and portrays in graphic detail whatone English Judge in a like context has called a ‘communism ofterror'. The whole nation was enveloped by a cloud of anxietyand fear”.
“It was anti-systemic i.e. violence directed to bring about thedestruction of the whole democratic structure of governmentand the collapse of all government institutions. It was amovement whose first aim was to annihilate the existing powerstructure and then attempt a revolutionary take-over of theapparatus of the State by armed force.”
If these passages are truly descriptive of what the evidenceshows, then it is reasonable to say that this climate of intimidation thatcommenced in 1987, continued from then on till the time of thegeneral election and beyond. I repeat that certain specific actsbrought out in the evidence do certainly fall within the description ofthose calculated to prevent the holding of the election. The evidenceof these acts apart, the rest of the evidence suggests not generalintimidation calculated to achieve the specific objective of preventing
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
99
the Presidential Election of 1988 from being held, but rather that thetrue picture is that this election was conducted while there waspresent throughout the country a climate of intimidation the object ofcreating which was to instil fear into the minds of the general public,voters and non-voters alike, so as to secure their submission andobedience to the demands and directions of this external force. If thismovement was "anti-systemic”, then the intimidation was calculatedto be against the system and all components of the system and theholding of an election being such a component was a target; but itwas not the sole target in the sense of a calculated effort to preventthe holding of the election isolated from the system. Indeed it must beremembered that 55% of the total number of registered voters,people of ordinary courage, men and women, young and old, went tothe polls openly and visibly and that to my mind suggests that theywere appreciative of the point made, that by and large anatmosphere of terror and tension had been created in the country forthe larger purpose of overturning the entire established system. A'statement to the effect then that this election was held during theprevalence of such a climate of intimidation does not necessarilymean that there was general intimidation as was altogethercalculated to prevent electors from voting or the majority of electorsfrom electing the candidate whom they preferred. The conclusion thatI draw then from the evidence taken as- a whole is, that althoughcertain aspects of such intimidation had an effect on the PresidentialElection of 1988, particularly such acts as were directed specificallyto discourage voters from voting, nonetheless one cannot say withany degree of certainty or reasonable probability that such electiondid not show a voter turn-out of 80% or indeed any otherascertainable percentage, as a consequence of general intimidationcalculated to prevent the holding of this election. When one is unableto draw a line of separation between the continued atmosphere ofintimidation and tension that prevailed in the country during this entireperiod and the acts of general intimidation committed which werecalculated to prevent the holding of the Presidential Election, one is inan uncertain area of speculation as to what percentage of loss ofvoter turn-out at such election, whether 25% as claimed or any otherquantifiable percentage, could be attributed to general intimidation,calculated to prevent electors from voting or the majority of electorsfrom electing the candidate whom they preferred. In this regard it is
100
Sri Lanka Law Reports
[1992] 2 Sri L.R.
also possible to say that though over 50% of ordinary persons ofaverage courage voted, others who might well have fallen within thatdescription though not intimidated, could nonetheless have desistedfrom voting considering it sensible to keep away from the poll, guidedby a counsel of prudence that discretion is the better part of valour,much in the same way that they might have been inclined to do, hadthey decided to remain in their homes without going about as freelyas they might have done under conditions of normalcy. The pointsought to be emphasized here is that quite a number of voters whomay have refrained from voting could well have done so as an act ofprudence and not because they were intimidated and thus it cannotbe said that they were prevented from voting, but rather that theyexercised a choice not to vote. Some sense of what I have in mindcan be got from the following words used by Mr. Justice Field in thecase of The Thornbury Division of the County of Gloucester(27) which Ihave already referred to “We must take the electors as an average ofordinary men who may be disinclined to go to the poll, but were notnecessarily intimidated”. How then can one apportion any precisepercentage of the drop in voter turn-out in December 1988 to generalintimidation as that expression is used in a context relevant to anelection, whether such percentage be 25 points or any other, as amatter of reasonable conclusion from the evidence in the case,without running the risk of committing a serious error? For myself, Ifind that I am unable to draw any reasonable conclusion, which theevidence would warrant as to what fraction of the percentage that didnot vote at the Presidential Election, as compared to the percentagethat might ordinarily have been expected in a normal situation tohave voted, was attributable to the practice of general intimidationcalculated to prevent voters from voting, which is what matters in thecase.
The next question that requires consideration is as to the majoritythe 1st respondent secured over the petitioner, which it was claimedwas a fraction of the order of 1/6th or 1/7th of the number ofadditional votes that would have come in, if 80% of the registeredvoters had voted. The following figures were submitted in support ofthe petitioner’s position that if this 80% did vote she could haveovertaken the 1st respondent by one vote and achieved victory if shehad obtained 279339 out of the additional votes numbering 2314371.
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
101
Total of Registered voters9375742
Total Polled (55,32%)5186223
If 80% voted the No. of votes that
would have been polled7500594
The No. of additional votes if 80%
polled 7500594 – 51862232314371
For the 1st Respondent to be declaredthe winner if 80% polled
He would have had to get 50% + 1 of
the 80% i.e. he must get3750298
He received2569199
The additional votes he would have
had to get 3750298 – 25691991181099
For the Petitioner to win if 80% polled
She too had to get 50% + 1 of the
80% i.e.3750298
She received2289860
The additional No. of votes would
be 3750298 – 22898601460438
The difference between
1460438 and 1181099 =279339
The number 279339 can be arrived at by deducting the total numberof votes the petitioner obtained from those that the 1st respondentdid, but a close look will show that the calculation relied on is itselfmade in such a way as to arithmetically achieve this result. At firstglance, these figures have no doubt the appearance of plausibility,but I will endeavour to demonstrate the erroneous basis upon whichthis result is reached. Section 56(2) of the Presidential Elections Actdirects The Commissioner of Elections, if he finds that any particularcandidate has received more than one half of the total number ofvalid votes cast, to forthwith declare such candidate elected. Ifhowever no candidate receives more than one half of the total valid
102
Sri Lanka Law Reports
[1992] 2 Sri L.R.
votes cast, then there is a procedure laid down which involves thecounting of preference votes. To achieve an 80% voter turn-out, anadditional 2314371 votes would have had to be cast. Upon thefigures presented for the petitioner, it was claimed that if 80% of allregistered voters did vote, the petitioner would have received onevote in excess of one half of the total number of votes cast andachieved victory, had she obtained an extra 1460438 votes inaddition to those she actually received at the election numbering2289860. Giving her that number of 1460438 votes out of theadditional 2314371 votes cast, would leave a balance of 853933votes. Now what happens to this remaining number of 853933 votes?They cannot be disregarded and must enter into the calculation. Letme assume that the entirety of that number is allotted to the 1strespondent. Then his total moves up to 3423132 which is a sum ofthat number plus 2569199, being the number of votes he actuallyreceived. It will be noted that this number 3423132 is less (bya figureof 327166) than the number 3750298 which is the total shown asnecessary for the petitioner to obtain one vote in excess of one half ofthe total of the votes cast and achieve victory. Why is that so? It is so,because out of the additional 2314371 votes taken into reckoning inthis exercise, to achieve her victory a percentage which works out toapproximately 63.10 has been allotted to the petitioner leaving abalance of only 36.90% available to be allotted to the 1st respondent,assuming that is, that the whole of it were to be allotted to him. If onthe other hand on the figures presented, the 1st respondent was toachieve victory having received one half of the total of the votespolled plus one out of the total number of 80% votes polled, acalculation on a like basis will demonstrate that he would have to beallotted approximately 51.03% of the extra 2314371 votes polled with48.97% available to be allotted to the petitioner. The result of thisexercise done in the manner claimed then, is to show that with an80% voter turn-out, to ensure the petitioner’s success she would havehad to obtain 63.10% out of the additional votes, being a percentagemuch in excess of the 44.95% that she actually received at theelection; while the 1st respondent’s share of the additional voteswould be reduced to 39.90% being much less than the 50.43% heactually received at the election. Similarly with a 80% voter turn-out,to achieve victory the 1st respondent would have had to obtain51.03% of the additional votes cast, being an increase marginally
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
103
over the 50.43% he actually received at the election, with 48.97%available to the petitioner out of the excess votes cast which onceagain is in excess of the 44.95% that she actually received at theelection. Therefore the picture that emerges on an 80% voter turn-outcalculated on the basis suggested is that the petitioner, whethersuccessful or unsuccessful, would have obtained a greaterpercentage of the extra votes cast than she actually received at theelection; whereas in the case of the 1st respondent if he wassuccessful he would have obtained a percentage of the extra votescast, marginally above the percentage he actually received at theelection, and if he was unsuccessful a percentage of the extra votescast, much lower than that which he received at the election.
To complete the picture, I will now take a situation where hundredpercent of the registered voters are considered to have voted anddeal with that situation in the manner in which the petitioner dealt withan 80% voter turn-out; that is in such a way as to maintain thearithmetical majority of 279339 which the 1st respondent actuallysecured over the petitioner at the election. To achieve such hundredpercent voter turn-out there should have been an additional 4189519votes. Half the total number of votes cast then would have been reallyhalf the total number of registered voters and would amount to4687871. The petitioner having received 2289860 votes, the balancerequired to make up one half of the total number of votes cast wouldhave been 2398011 votes with an additional one vote necessary toachieve victory bringing the number up to 2398012. That wouldrepresent a 57.23% of the extra votes being allocated to her with thebalance 42.77% available to the 1st respondent. If on the other handthe 1st respondent was to have achieved victory, in addition to the2569199 votes he received, he would have had to obtain anadditional 2118672 out of the extra votes to reach the level of one halfof the votes cast with one vote added to achieve victory, that numberthen becoming 2118673. That would be 50.57% out of the extra voteswith 49.43% available to the petitioner.
This exercise demonstrates the anomalies that arise whenhypothetical figures are taken as constituting total voter turn-outpercentages and it is sought to maintain the figure of 279339 beingthe majority the 1st respondent in fact received over the petitioner atthe election. In doing so one sees that varying hypothetical
104
Sri Lanka Law Reports
[1992] 2 Sri LR.
percentages of the extra votes cast have to be assigned to theparties, percentages completely different from those actually polledat the election with no basis, factual or arithmetical to justify doing so.
A more realistic method that commends itself to me is to base apossible calculation on the actual votes cast. The valid votes polledat the election were thus:-
Oswin Abeygunasekera235719
The petitioner2289860
The 1st respondent2569199
The total number therefore of
valid votes cast5094778
Now the effect of what was contended for the petitioner is that toequal the 1st respondent’s total of 2569199, all that was necessarywas to obtain 279339 extra votes and that one more vote beyond thatnumber would give her a majority as well as success at the election.Arithmetically it would be correct to say that she would have receivedsuch majority, assuming that the 1st respondent’s number of votesremained static. By receiving this number of votes the petitionerwould have overtaken the 1st respondent’s total by one vote. Butadding 279340 votes to the votes actually polled by the petitionerwould mean that the total number of votes cast in the election wouldhave increased by that number because such number could nothave been taken out of the votes cast for either of the othercandidates. When the total votes cast gets increased by 279340adding up to the number 5374118, one half of that number becomes2687059 and the petitioner’s total would have been what shereceived, namely 2289860 plus the 279340 added for the purpose ofthis exercise, totalling to 2569220. This figure falls short of the level ofone half of the total votes cast required and upon that total, she couldnot have been declared the winning candidate under section 56(2) ofthe Presidential Elections Act. Nor for that matter, would the 1strespondent have received a figure in excess of one half of the totalthus made up, and this would have resulted in the need to count thepreference votes in order to ascertain who the winning candidatewould have been. Arithmetically it will then be seen that in order to
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
105
reach the level of one half of the total votes cast, the petitioner wouldhave had to receive in addition to the number she actually polled, anextra 515058 votes and to exceed that number, another one vote.That too would be on the basis that neither of the other candidatesreceived a single extra vote apart from what they actually polled. Tomaintain that one vote lead and at the same time to maintain the levelof having reached one half of the total of the votes cast, for each onevote either of the other candidates received, the petitioner herselfwould have had to receive an additional countervailing vote. Themanner in which the number 515058 was reached is thus. The votesreceived by the 1st respondent numbering 2569199 and Mr. OswinAbeygunasekera numbering 235719 total to 2804918. The petitionerhaving received 2289860 could have equalled that total by obtaining515058 extra votes and with a single vote added to that number it willbe seen that while achieving victory she would have also reached thelevel of having secured one half of the total votes cast.
As I see it therefore, it is not accurate for the petitioner to contendin the context of the Presidential Elections Act that the majority whichshe had to overcome was 279339 votes, which therefore constitutedonly a 1/6th or 1/7th of the total votes that would have been polledhad 80% of the voters in fact voted at the election, and that inconsequence the majority of electors may have been prevented fromelecting the candidate whom they preferred in quite the manner aswas contended for her.
Apart from this there is another aspect which cannot bedisregarded. One must take care to be reminded of the fact that thisproposition as contended for the petitioner is based upon the actualfigures that became available at the conclusion of the election. Theproposition is that the 1st respondent’s majority over the petitioner'stotal of votes was only 1/6th or 1/7th of the 25% shortfall in the votes.It is with respect to such a situation that it was contended for thepetitioner, that if she secured a further 279339 votes she could haveerased the 1st respondent’s majority over her. But the question is, ifthe extra 25% did vote, who is to know what the distribution patternof these votes between the three candidates would have been, so asto enable one to say, as claimed, that if the petitioner obtained279339 votes in addition to those she actually received, she would
106
Sri Lanka Law Reports
[1992] 2 Sri L.R.
have overtaken the 1st respondent’s majority and at the same timereceived one half of the total number of votes cast. The system ofpreference votes could also, in certain circumstances, be imaginedto have caused yet another complication, if it had become necessaryto take such preference votes into account by reason of an increasedvoter turn-out, as a consequence of no candidate having receivedover one half of the total votes polled. In that situation if the candidatewho received the lowest number of votes had to be eliminated, hisvotes would then have had to be distributed among the othercandidates according to the next preference indicated by his voters.In what proportion they would have been distributed would be anunknown factor. Without knowing that, one would once again not bein a position to say, what the difference would have been in thenumber of votes between those secured by the winning candidateand by the runner-up, after those votes were taken into reckoning. Ina situation therefore where 80% of the voters might have beenexpected to vote, how the extra 25% might in fact have voted notbeing known, one cannot say whether or not the preference votes ofthose who voted for the third candidate would have had to becounted and then one cannot say except as a matter of guessworkwhat the difference in votes would have been between those of thewinning candidate and the runner-up.
The essence of Mr. H. L. de Silva’s contention with respect to theorder on the preliminary objections was, that in interpreting themeaning of the phrase, “the majority of electors may have beenprevented from electing the candidate whom they preferred", theCourt, as I earlier said, used the word “result”, not in the sense of anumerical count of votes, but in the sense of a consequence or effecton the electors’ choice of candidate. This, he said, was explained inthe very next sentence the Court used when it said, “But it will beopen to the returned candidate to show the gross intimidation couldnot possibly have affected the result of the election.” His submissiontherefore was, that if the number of voters who had been preventedfrom voting by reason of the gross intimidation had been six or seventimes as great as the majority the declared candidate secured overthe next, then there is demonstrated that there had been such aconsequence or effect on the electors’ choice of candidate, as wouldenable one to say that the majority of voters may have been
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
107
prevented from electing the candidate whom they preferred. Hisargument therefore was, that in order to dispel the possibility sosuggested that the majority of electors may have been preventedfrom electing the candidate whom they preferred, it becameincumbent upon the returned candidate to show that the shortfall inthe votes due to the gross intimidation was so small as to be lessthan the majority of the returned candidate, by which process thereturned candidate would then have demonstrated that the grossintimidation could not possibly have affected the result of theelection. There are certain matters that need to be considered in thisregard. There is first Mr. H. L. de Silva’s contention as to the degree ofproof required, to discharge the burden that lies upon the petitionerwhen invoking the provisions of section 91(a) of the PresidentialElections Act to have the election declared void. The section reads tobe understood that, even if it may not be proved that the majoritywere in fact so prevented, it would be sufficient to prove that themajority may have been so prevented. Mr. H. L. de Silva argued thatthe use of the word ''may" in the section indicates that what wasrequired was to show that there was that “possibility” and that such aburden therefore would be discharged upon a lower standard ofproof than suggested by the word “probability". The word “may” asused in the section, he submitted, was equivalent in meaning to theuse of the word “possible,” used as the antonym of the word“impossible". The line of argument was that in a situation where aparticular thing is “possible", equally that same thing could be, “notpossible" and that these two situations could co-exist without being inconflict with each other, so that to exclude a thing being “possible”, itwas necessary to prove that such thing was "impossible”. Counselargued that if the evidence indicated that the winning candidate’smajority was 1/6th or 1/7th of the number that constituted 25% of thetotal of registered voters and that they were so prevented from voting,there was an “inference" or “presumption” that arose, that the majorityof electors may have been prevented from electing the candidatewhom they preferred. If that is the situation contemplated by theword, “may”, as meaning, “possible", the question that occurs to meis whether there could not be the same kind of possibility as ameasure of the degree of proof required, in a situation where themajority secured by the winning candidate, instead of being 1/6th or1/7th, is just short of the shortfall in votes. To displace such inference
108
Sri Lanka Law Reports
[1992] 2 Sri LR.
or presumption, there was, he submitted, a burden cast on the 1strespondent according to the Court's order on the preliminaryobjections, to show that the majority of electors were, in fact, notprevented from electing the candidate whom they preferred, aswould be the case in a situation where the number of those who wereprevented from voting was less than the winning candidate's majority.What he submitted amounts to this, that the petitioner haddischarged the burden which lay upon her by virtue of such inferenceor presumption, and that consequently the burden had shifted to the1st respondent to show that the majority of electors were notprevented from electing the candidate whom they preferred. If it becorrect that such a burden came to be cast upon the 1st respondent,what in reality is the effect of his so demonstrating in that manner thatthe majority of electors had not been prevented from electing thecandidate whom they preferred. Then a numerical test is used torebut a hypothetical possibility. The moment such a thing is done isthere not implicit in the exercise, an endeavour at discoveringwhether another candidate should have achieved victory instead ofthe returned candidate? The effect of doing that is, as I see it, that byan exercise involving a numerical reckoning of votes, it is sought toestablish that the declared candidate's success remains unaffected.To show that in that way, is to show that the result of the election wasnot affected, using that expression in a sense analogous to that inwhich its words are used in section 91 (b) of the PresidentialElections Act. Then, one is in the curious position of saying that the1st respondent is called upon to demonstrate numerically that theresult of the election was not affected, to meet a case upon which thepetitioner herself is not called upon to demonstrate numerically thatthe result of the election was affected.
Something else needs to be said about the contention of learnedCounsel, that it is sufficient to show the existence of a mere possibilitythat the majority of electors may have been prevented from electingthe candidate whom they preferred. The section also deals with asituation of proof, to the satisfaction of the Court, that the majority ofelectors were in fact prevented from electing the candidate whomthey preferred. This, as Mr. H. L. de Silva contended, is a degree ofproof which demonstrates a measure of certainty. If then that be so,the degrees of proof covered by the section would span the whole
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
109
range, starting from a level of certainty at one end and extending to alevel of a mere possibility or anything other than impossibility at theother, a situation which it does not seem at all likely that thelegislature could be thought to have intended as one reads andunderstands the section.
Much reliance was placed for the petitioner on the judgment of theCourt of Common Pleas in the case of Woodward v. Sarsonsm. It wascontended that, not merely does the judgment set out, correctly andfully, the common law concept of a free and fair election, but also thatit was the inspiration that guided, both in language and in concept,the entry into our law of the very first section in terms similar tosection 91(a) of the Presidential Elections Act. That case arose out ofan allegation that there was non-compliance with the rules and formsprovided for the holding of elections. The judgment of the Court(consisting of Brett, Archibald and Denmen JJ) read by Coleridge,C.J. contains (at pages 743 and 744) the following passages:
‘The questions raised for decision seem to be. First, what isthe true statement of the rule under which an election may beavoided by the common law of Parliament? secondly, is thepresent case brought within the rule? …
As to the first, we are of opinion that the true statement, isthat an election is to be declared void by the common lawapplicable to parliamentary elections, if it is so conducted thatthe tribunal which is asked to avoid it is satisfied as a matter offact, either that there was no real electing at all, or that theelection was not really conducted under the subsisting electionlaws. As to the first, the tribunal should be so satisfied, i.e. thatthere was no real electing by the constituency at all, if it wereproved to its satisfaction that the constituency had not in facthad a fair and free opportunity of electing the candidate whichthe majority might prefer. This would certainly be so, if a majorityof the electors were proved to have been prevented fromrecording their votes effectively according to their own
110
Sri Lanka Law Reports
[1992] 2 Sri LR.
preference, by general corruption or general intimidation, or bybeing prevented from voting by want of the necessarymachinery for so voting, as, by polling stations beingdemolished, or not opened, or by other of the means of votingaccording to law not being supplied or supplied with sucherrors as to render the voting by means of them void, or byfraudulent counting of votes or false declaration of numbers bya returning officer, or by such other acts or mishaps. And wethink the same result should follow if by reason of any such orsimilar mishaps, the tribunal, without being able to say that amajority had been prevented, should be satisfied that there wasreasonable ground to believe that a majority of the electors mayhave been prevented from electing the candidate theypreferred. But, if the tribunal should only be satisfied that certainof such mishaps had occurred but should not be satisfied eitherthat a majority had been, or that there was reasonable ground tobelieve that a majority might have been, prevented fromelecting the candidate they preferred, then we think that theexistence of such mishaps would not enable the tribunal todeclare an election void by the common law of Parliament. Thiswe think, is the result of comparing the judgments of Grove, J.,at Hackney™ and Dudleym with the judgment of Martin, B., atSalford™ and of Mellor, J., at Bolton™ all of which judgmentsare in accordance with, but express more accurately, thegrounds of the decisions in parliament in the older cases ofNorfolk(34), Heyw, Co. 555.(n)(3S), Morepetht3B), Pontefract(37>,Coventry<38>, New Rossm, & Drogheda™ & the Droghedacase™ all of which are mentioned in Rogers on Elections, 10thed, 365 et seq.
As to the second, i.e. that the election was not reallyconducted under the subsisting election laws at all, we think,though there was an election in the sense of there having beena selection by the will of the constituency, that the question mustin like manner be, whether the departure from the prescribedmethod of election is so great that the tribunal is satisfied, as amatter of fact, that the election was not an election under the
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
111
existing law. It is not enough to say that great mistakes weremade in carrying out the election under those laws: it isnecessary to be able to say that, either wilfully or erroneouslythe election was not carried out under those laws, but undersome other method.”
One sees that the Court of Common Pleas considered that atribunal examining the question as to whether an election should bedeclared void by the common law, is called upon to be satisfied thatthere is reasonable ground to believe that the majority of electorsmay have been prevented from electing the candidate they preferred,in a situation where it is not proved that such majority of electors werein fact prevented from electing the candidate whom they preferred.Now what do the words "reasonable ground to believe”, as used bythe judges in that case mean? Do they or do they not have the samemeaning as that which Mr. H. L. de Silva contended for, with respectto section 91(a) of the Presidential Elections Act. His response, inanswer to a query from the Court as to this was, that the probativefacts or the evidentiary facts have to be established upon a balanceof probability, but that the factum to be proved is still in regard to thepossibility of an event or an occurrence and not as to the probabilityof its occurrence. I must confess that I did not quite comprehend theintended meaning of this submission. Assuming that he meant by thisthat it had to be established upon a balance of probability, that therewas a degree of general intimidation as resulted in a shortfall in voterturn-out of 25% and that there was also a probability that the majoritysecured by the winning candidate was as low as 1/6th or 1/7th of thatshortfall, is it that there arises a possibility of an event or occurrenceand if so, is that event or occurrence the prevention of the majority ofelectors from electing the candidate whom they preferred? That isthe only factum to be proved that I can see upon this contention, thatthe majority of electors may have been prevented from electing thecandidate whom they preferred, a factum to be proved as a merepossibility. If that be the factum to be proved as a mere possibility,that the majority of electors may have been prevented from electingthe candidate whom they preferred, then the factum to be provedaccording to the judgment in Woodward v. Sarsons is also that themajority of electors may have been prevented from electing thecandidate whom they preferred, but there, as set out by the Judges,as a reasonable ground of belief, which is certainly not to say as a
112
Sri Lanka Law Reports
[1992] 2 Sri L.R.
matter of mere possibility. If therefore the judgment in Woodward v.Sarsons as to the common law is embodied in section 91(a) of thePresidential Elections Act. Mr. H. L. de Silva’s submission is one I findthat I cannot assent to. I rather consider that the correct formulation isthat which the judges in Woodward v. Sarsons adopted, upon whichit would not suffice to say that the petitioner could discharge herburden by pointing to a mere possibility (as opposed to animpossibility) that the majority of electors may have been preventedfrom electing the candidate whom they preferred, but that she hasindeed to satisfy the Court as a reasonable ground of belief that themajority of electors may have been prevented from electing thecandidate whom they preferred, which in my understanding meansthe same thing as saying, furnishing proof as a matter not even ofmere probability, but indeed of reasonable probability.
It now becomes necessary to say that there was a burden castupon the petitioner to show that the majority of electors may havebeen prevented from electing the candidate whom they preferred,whichever meaning that expression has, which is to show that there isreasonable ground to believe that such majority were so prevented,and that renders it necessary to look at the provisions of the EvidenceOrdinance relating to the burden of proof in this case. As Mr. Choksypointed out, section 91(a) of the Presidential Elections Act requiresthat it must be proved to the satisfaction of the Court, that by reasonof general intimidation the majority of electors were in fact prevented,or that the majority of electors may have been prevented fromelecting the candidate whom they preferred. As emphasized byMr. Choksy, in terms of section 3 of the Evidence Ordinance, a fact issaid to be proved, when after considering the matters before it, theCourt either believes it to exist, or, considers its existence soprobable that a prudent man ought in the circumstances of theparticular case, to act upon the supposition that it exists. If thatdefinition of proof is applied to the terms of section 91(a) of thePresidential Elections Act, when an election is sought to be avoidedon the basis of general intimidation, the position would be thus,whichever meaning one takes to be the true meaning of theexpression, “the majority of electors were or may have beenprevented from electing the candidate whom they preferred”.
sc
Sfrimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonev/ardene, J.)
113
On the basis that the majority of electors were in fact preventedfrom electing the candidate whom they preferred, (a) the Court mustbelieve as a matter of actual existence that the majority of electorswere prevented from electing the candidate whom they preferred, or,(b) the Court must consider that under the circumstances of thecase, a prudent man ought to act on the supposition that the majorityof electors were prevented from electing the candidate whom theypreferred.
On the basis that the majority of electors may have beenprevented from electing the candidate whom they preferred, (a) theCourt must believe as a matter of actual existence that the majority ofelectors may have been prevented from electing the candidate whomthey preferred or, (b) the Court must consider that under thecircumstances of the case, a prudent man ought to act on thesupposition that the majority of electors may have been preventedfrom electing the candidate whom they preferred.
The next point to note, and one of importance to the result of thiscase is as to the provisions of the Evidence Ordinance on thequestion of burden of proof. They do have the effect of casting theburden of proof in the case upon the petitioner to establish all thenecessary ingredients which would entitle her to the relief shedemands. Sections 101, 102, and 103 may usefully be reproducedhere.
Section 101. " Whoever desires any court to give judgmentas to any legal right or liability dependent on the existence of factswhich he asserts, must prove that those facts exist.
When a person is bound to prove the existence of any fact, itis said that the burden of proof lies on that person.”
Section 102. “ The burden of proof in a suit or proceeding lieson that person who would fail if no evidence at all were given oneither side.”
Section 103. “ The burden of proof as to any particular factlies on that person who wishes the Court to believe in its existence,
114
Sri Lanka Law Reports
[1992] 2 Sri L.R.
unless it is provided by any law that the proof of that fact shall lie
on any particular person.”
The burden then lay on the petitioner to show, not as a matter ofmere probability but as a matter of reasonable probability, that themajority of electors were or may have been prevented from electingthe candidate whom they preferred, by reason of generalintimidation. It is my belief that the order of the Court which Mr. H. L.de Silva said was to cast a burden on the 1st respondent, really didintend to say that the petitioner had to furnish the requisite proof as tothis as a reasonable ground of belief, but that it was open to the 1strespondent to show upon a balance of probability, that a view of thewhole of the evidence was one more favourable to him.
Upon the hypothesis advanced by learned Counsel for thepetitioner as to the meaning to be gathered from the expression, “ themajority of electors may have been prevented from electing thecandidate whom they preferred”, which he submitted found favourwith the Court as shown by its order on the preliminary objections, Ihave examined the ingredients he referred to as necessary to beproved by the petitioner. Upon such hypothesis and on her owncontention as to what she had to establish, I have to conclude as amatter of reasonable probability or reasonable ground of belief, thatthe petitioner has failed to establish, the following ingredients whichare, that a particular percentage of electors refrained from voting byreason of general intimidation calculated to prevent them from doingso, that the number of additional votes sufficient to entitle her tosuccess was the number she relied upon which therefore was afraction of the order of 1/6th or 1/7th of the shortfall, and that therewas thus demonstrated that the majority of electors may have beenprevented from electing the candidate whom they preferred in thesense contended on her behalf. The reasons for saying this I havealready explained with respect to each ingredient separately andthus I cannot accept that the combined effect of the material relied onto support each component was to bring about the consequencecontended for the petitioner. The conclusion then must be, that uponthe formulation adopted by learned Counsel for the petitioner, hercase presented on the basis of general intimidation must fail.
sc
Sirimavo Bandaranalke v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
115
That should perhaps ordinarily have sufficed to dispose of thematter, that being on the basis of the case presented for thepetitioner, but there yet remains not ascertained, the true meaning ofthe expression, “the majority of electors were or may have beenprevented from electing the candidate whom they preferred”. I willtherefore proceed to address my mind to that.
Usefully, in approaching the questions then involved, I would startwith the formulation adopted by the Court of Common Pleas in thecase of Woodward v. Sarsons. From the passages of the judgmentwhich I have already reproduced, the following propositions asrespects general intimidation can be deduced.
An election is to be declared void by the common law, if it was soconducted that the tribunal is satisfied as a matter of fact, that therewas no real electing by the constituency at all.
The tribunal should be satisfied that there was no real electing bythe constituency at all, if it were proved that such constituency hadnot, in fact, had a fair and free opportunity of electing the candidatewhich the majority of the electors might prefer.
3..The constituency would not in fact have had a fair and freeopportunity of electing the candidate which the majority of theelectors might prefer if a majority of the electors were proved to have. been prevented by general intimidation from recording their voteseffectively according to their own preference.
The same result should follow (that is that the election should bedeclared void), if the tribunal, without being able to say that a majorityhad been prevented, should be satisfied that there was reasonableground to believe that a majority of the electors may have beenprevented by general intimidation from electing the candidate theypreferred.
Even if the tribunal should be satisfied that some generalintimidation had taken place, .but nevertheless should not be satisfiedthat a majority of the electors had been prevented from electing thecandidate they preferred, the tribunal would not be entitled to declare
116
Sri Lanka Law Reports
[1992] 2 Sri LR.
the election void by the common law, by reason only of the existenceof such general intimidation.
Even if the tribunal should be satisfied that some generalintimidation had taken place, but nevertheless should not be satisfiedthat there was reasonable ground to believe that a majority of theelectors had been prevented from electing the candidate theypreferred, the tribunal would not be entitled to declare the electionvoid by the common law, by reason only of the existence of suchgeneral intimidation.
In order to ensure accuracy, I have endeavoured to extract thesepropositions from the judgment with care. If I am to rearrange theideas expressed in the propositions numbered 1, 2 and 3 in thereverse order so as to see how they would read, this is the result. Ifthe majority of electors were proved to have been prevented bygeneral intimidation from recording their votes effectively accordingto their own preference, (which is to say they were prevented fromvoting as they wished) the constituency would not have had a fairand free opportunity of electing the candidate whom they preferred. Ifthe constituency did not have a fair and free opportunity of electingthe candidate whom they preferred, there would be no real electingby the constituency at all. If there was no real electing by theconstituency at all, the election is to be declared void. If oneexamines the proposition numbered 3 alone, it will be seen that itappears to point to a situation as to when it could be said that the .majority of the electors have not had a fair and free opportunity ofelecting the candidate whom they preferred. That is to be found inthe words, “a majority of the electors . . . were prevented fromrecording their votes effectively according to their own preference". Ifthe majority of electors were prevented from voting as they wished,there could then be no doubt that the election must be declared void.Let me now see what the situation is, when it cannot be said that themajority of electors were prevented from electing the candidatewhom they preferred, but it is possible to say there are reasonablegrounds to believe that the majority of electors may have beenprevented from electing the candidate whom they preferred. In termsof the proposition numbered 4 above, then too the election must bedeclared void. If however the majority of the electors were not
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
117
prevented from recording their votes effectively according to theirown preference, what follows? It is perhaps on the basis that it couldthen be said that the constituency was not deprived of and thereforedid have a fair and free opportunity of electing the candidate whichthe majority of the electors might prefer that a position appears tohave been taken for the respondents at an early stage in the case,that the charges based upon section 91(a) of the PresidentialElections Act must fail, because in the event, the majority of theelectors (over 50% of the registered voters) had the opportunity anddid in fact vote. I however mention this only in passing, but make nopoint of it.
Since the occasion lends itself to such a step, I would refer here tothe aspect of the petitioner’s case as pleaded in her petition inparagraph 6C based upon the provisions of section 91(a) of thePresidential Elections Act as against the 2nd respondent, on theground of non-compliance with the provisions of the Act relating tothe conducting of the election in accordance with the principlescontained in such provisions. The case of Woodward v. Sarsons isimportant in this regard as well, and the several examples of votersbeing prevented from voting by want of the machinery necessary forso voting as shown in the quotation reproduced, are alsocharacterised as instances demonstrating that there was no realelecting by the constituency at all. A set of like propositions asrespects that aspect is once again extracted from the judgmentthus:-
An election is to be declared void by the common law, if it was soconducted that the tribunal is satisfied as a matter of fact, that therewas no real electing by the constituency at all.
The tribunal should be satisfied that there was no real electing bythe constituency at all, if it were proved that such constituency hadnot, in fact, had a fair and free opportunity of electing the candidatewhich the majority of the electors might prefer.
The constituency would not in fact have had a fair and freeopportunity of electing the candidate which the majority of theelectors might prefer if a majority of the electors were proved to have
118
Sri Lanka Law Reports
[1992] 2 Sri L.R.
been prevented by want of the machinery necessary for so voting,from recording their votes effectively according to their ownpreference.
The-same result should follow (That is that the election should bedeclared void), if the tribunal, without being able to say that a majorityhad been prevented, should be satisfied that there was reasonableground to believe that a majority of the electors may have beenprevented by want of the machinery necessary for voting, fromelecting the candidate they preferred.
Even if the tribunal should be satisfied that there was somebreakdown of the machinery necessary for voting, but neverthelessshould not be satisfied that a majority of the electors had beenprevented from electing the candidate they preferred, the tribunalwould not be entitled to declare the election void by the common law,by reason only of the existence of those instances of breakdown ofthe machinery necessary for voting.
Even if the tribunal should be satisfied that there was somebreakdown of the machinery necessary for voting, but neverthelessshould not be satisfied that there was reasonable ground to believethat a majority of the electors had been prevented from electing thecandidate they preferred, the tribunal would not be entitled to declarethe election void by the common law, by reason only of the existenceof those instances of breakdown of the machinery necessary forvoting.
Thus, I think it can be said that, as the judgment in Woodward v.Sarsons shows, the English common law concept of a free and fairelection has relevance not only when considering the element ofgeneral intimidation, but also in considering the element of failure toconduct an election, with the necessary machinery provided forvoters to vote, in either case, so as to prevent the majority of electorsfrom electing the candidate whom they preferred.
If that be a true statement of the common law relating to theconcept of a free and fair election as it was understood in England,let me now examine our statutory provision vis-a-vis the judgment of
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
119
the Court of Common Pleas in Woodward v. Sarsons. It must be keptin mind that the relevant section, namely, section 91(a) of thePresidential Elections Act, is a reproduction of earlier legislation insimilar terms, the earliest to be found, if I understood Mr. H. L. deSilva correctly, in the 1931 State Council Elections Order-in-Council.
Putting together the relevant parts of the main section 91 and ofsection 91 (a), the provision may be made to read thus:-
The election . . . shall be declared to be void … on … thefollowing ground(s) which may be proved …. namely, that by reasonof general intimidation … the majority of electors were or may havebeen prevented from electing the candidate whom they preferred.
A rearrangement of these clauses to get out the sense moreclearly would, as I see it, read:
The election shall be declared to be void on the followingground(s) which may be proved, namely, that the majority of electorswere or may have been prevented from electing the candidate whomthey preferred, by reason of general intimidation.
When this section is rendered in this form it helps to identify apossible misconception that can occur, of much significance in thecase. It is seen at once that the ground of avoidance is that themajority of electors were or may have been prevented from electingthe candidate whom they preferred. One can all too easily fall into theerror of thinking that a ground of avoidance under the section isgeneral intimidation. The ground of avoidance under section 91(a)then is really the effect, which is that the majority of electors were ormay have been prevented from electing the candidate whom theypreferred and not the cause, which may be general intimidation.
The Court of Common Pleas considered in the case of Woodwardv. Sarsons that the concept of a free and fair election (or an electionat which there was a real electing) implies one at which the majorityof electors had the opportunity of electing the candidate whom theypreferred. If they therefore, that is such majority, did not have thatopportunity, then there was no free and fair election or no real
120
Sri Lanka Law Reports
[1992] 2 Sri L.R.
electing, to state the other side of the proposition, the consequencethen being that the election must be declared void.
What then is the true meaning of the language used in section91(a) which is said to have derived its inspiration from the judgmentin this case? The answer to that is made simpler I think by the way Ihave rendered the section. The election shall be declared void whena certain eventuality or situation exists. What is that eventuality orsituation? It clearly is that the majority of electors were or may havebeen prevented from electing the candidate whom they preferred, asthe language used reads. Prevented how, or in what way? Preventedby reason of general intimidation or failure to comply with theelections law, as the case may be, that being therefore the agency bywhich such prevention was brought about. The identical positionarises from the judgment'in the case of Woodward v. Sarsons, as thepropositions I have extracted show.
Such is the way I think the sense of the section must be got,because any true reading of it must give effect to all its words andclauses as may not be the case if the section is read to understandas if the ground of avoidance is general intimidation.
Something that strikes me as being supportive of the assertion Imake, that what is vitally important in the section are the wordssuggesting the need to establish that the majority of electors were ormay have been prevented from electing the candidate whom they •preferred is this. Reading the section (erroneously) as if the ground ofavoidance is failure of the electoral machinery, when that"circumstance” is relied on, there is no criterion upon which theelection could be avoided.
Summarising what I say in this regard, in terms of section 91(a) ofthe Presidential Elections Act, in order to avoid an election it must beshown that it was not a free and fair one. It is proved not to be a freeand fair election, when it is proved that the majority of the electorswere or may have been prevented from electing the candidate whomthey preferred (see here for comparison the propositions extractedfrom the judgment in Woodward v. Sarsons). That which prevents anelection from being a free and fair one may be, so far as is relevant to
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
121
this case, either the existence of general intimidation or the failure tocomply with the elections law. What may be thought of as the causewhich could be the presence of general bribery, general treating,general intimidation, misconduct or other circumstance whethersimilar to these or not, must be demonstrated to be present andshown to bring about the stated effect, which is that the majority ofvoters were or may have been prevented from electing the candidatewhom they preferred. Then only can the desired consequence beachieved, which is to have the election avoided. The cause cannotachieve the consequence without the effect being shown. It istherefore, in my view, vital to the success of the petitioner’s case asbased upon section 91(a) of the Presidential Elections Act to proveas the primary requisite, that the majority of electors were or mayhave been prevented from electing the candidate whom theypreferred. If it cannot be proved that the majority of the electors wereor may have been prevented from electing the candidate whom theypreferred, then the very ground relied upon, the essential factumprobandum, is not proved and it cannot then be said that there wasno free and fair election, whether the basis be general intimidation ornon-compliance with the elections law or any of the other factorsreferred to in the section, and the election therefore cannot beavoided. The concept of a free and fair election is not one to befound by reference to the words “general intimidation”, but byreference to the provision, "that the majority of electors were or mayhave been prevented from electing the candidate whom theypreferred"
I will now attempt an analytical approach to the question as towhat the words of this expression mean. The section contemplatestwo situations. The first is one where the majority of electors were (infact) prevented from electing the candidate whom they preferred, asituation of the highest importance in my view to the ascertainment ofthe true meaning of this expression. This position as requiring anelement of certainty by way of proof is reflected in the propositionsnumbered 1 to 3 relating to general intimidation which I haveextracted from the judgment in the case of Woodward v. Sarsons. Thelegislature when dealing with that situation, spells out a test in section91(a) and, independent of any question whether what is implicit inthat test is capable of proof or not, such test as a matter of clearlanguage is this. Were the majority of electors prevented from
122
Sri Lanka Law Reports
[1992] 2 Sri L.R.
electing the candidate whom they preferred? If the answer to that isin the affirmative, the next question that arises from it is whether itwas general intimidation that brought about that effect, the twoquestions having to be looked at in that order. Let me for the momentview this expression in the background of the present case. If, evenin the absence of the general intimidation (or for that matter acombination of the factors of general intimidation and a breakdown ofthe machinery of voting), it was thelst respondent who would yethave been the successful candidate, then it could not be said thatthe majority of electors were prevented from electing the candidatewhom they preferred, inasmuch as, with or without such generalintimidation, thelst respondent would have been the preferredcandidate of the majority of electors. Logically then, to be able to saythat the majority of electors were prevented from electing thecandidate whom they preferred, one must be in a position to say thatit was someone other than thelst respondent who would have beenthe successful candidate whose election by the majority of electorswas prevented by general intimidation, the vitiating factor, being theexpression used by Mr. H. L. de Silva. Viewed from another angle, themajority of electors at this election were not prevented from voting orexercising their franchise, as over 50% did vote. That majority thatdid vote, were once again not prevented from electing the candidatethey preferred. The question then is as to the minority who may havebeen prevented from voting or exercising the franchise. What wouldhave been the position, had such minority been able to contributetheir votes to the total polled? If their contribution was not to changethe result as to who would have won the election, as a matter of legalconsequence, the election must be sustained. If on the other handtheir contribution was to change the result of the election, as a matterof legal consequence, the election must be avoided. Then, whetherthe election must be avoided or not, depends on whether there wouldhave been a change in the successful candidate, had the minority ofelectors who were prevented from voting in fact voted, which thenimplies a change in the result, meaning that the result of the electionwas affected.
As I have pointed out, to be able to say that the majority of electorswere prevented from electing the candidate whom they preferred,one must be in a position to say, that it was someone other than the1st respondent who would have been the successful candidate
sc
Sirimavo Bandaranalke v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
123
whose election by the majority of electors was prevented by thevitiating factor. It does mean then, that the expression was calculatedto deal with a situation showing that someone else other than thedeclared candidate would have succeeded when the expression "themajority of electors were (in fact) prevented from electing thecandidate whom they preferred”. If that is so with respect to asituation of certainty, that is, where the majority were in factprevented, a situation contemplated by the words “the majority ofelectors were prevented from electing the candidate whom theyprefer", what is the position when using the expression “the majorityof electors may have been prevented”? One must necessarilyattribute the same meaning to the words, “majority of electors . . .prevented from electing the candidate whom they preferred”, whetherthe gap I have left in the sentence is to be filled by the word “were”on the one hand or by the words "may have been”, on the other; elsethere is something very wrong in the way the expression has beenstructured, and I, for myself, do not see anything wrong in thatregard. It deals with, as I see, a situation that had been contemplatedby the judgment in the case of Woodward v. Sarsons and shown byme as proposition numbered 4 above, in those relating to generalintimidation, where the Court is faced with a situation contemplatedby the words in the judgment "the tribunal without being able to saythat a majority had been prevented should be satisfied that there wasreasonable ground to believe that a majority of electors may havebeen prevented”. When encountering such a situation, the Court ofCommon Pleas has indicated what the proper approach should be,and that approach is not to give a different meaning to the words inthe two different contexts of, “were”, and, “may have been”, but toprovide an index as to the degree of proof then demanded which is"reasonable ground to believe” which again means, in myunderstanding, as a ground of reasonable probability.
If the Court is in a position to say, having reference to the facts ofany case, that the majority of electors may have been prevented fromelecting the candidate whom they preferred, as a reasonable groundof belief or as a reasonable probability, the Court it must be notedwould more readily be able to say the same when there is proof thatthe majority of electors were (in fact) prevented from electing thecandidate whom they preferred, as a matter of certainty. That is to
124
Sri Lanka Law Reports
[1992] 2 Sri LR.
say that if the legislature used only the wider expression, “the majorityof electors may have been prevented from electing the candidatewhom they preferred", such expression would have been adequateto encompass the narrower situation, where the majority of electors-were (in fact) prevented from electing the candidate whom theypreferred. Why then did the legislature use words to include thatsituation as well? That perhaps was done in order to give clarity to theexpression so that the words “the majority of electors may have beenprevented from electing the candidate whom they preferred” couldget their colouration from the words “the majority of electors wereprevented from electing the candidate whom they preferred", so as todenote a difference in the result if any, in the sense of anothercandidate’s success, much the same way the Court of CommonPleas gave expression to this notion in its judgment in the case ofWoodward v. Sarsons.
If one looks at some of the early English cases, notably thosedecided before the secrecy of voting provision had been introducedinto the law, one may get the impression that even a small degree ofgeneral intimidation, whatever its effect on the result may have been(even if negligible), was considered to be sufficient to avoid anelection (see for example the response of Mr. Justice Keogh to thearguments of Counsel in the Drogheda case that I reproducedearlier). Such an impression would be on much the same lines as thatsuggested by erroneously considering in the manner I have alreadypointed out, that under section 91(a) of the Presidential Elections Actthe ground of avoidance reads to be understood to be generalintimidation instead of the prevention of the majority of electors fromelecting the candidate whom they preferred. Mr. H. L. de Silva’sstrong reliance upon some of the passages in the older cases onthese lines, leads me to believe that he was himself contending forsuch a view. Such an impression though, is not that which is reflectedas the effect of the cases as the Court of Common Pleas saw it in thecase of Woodward v. Sarsons. Even if the emphasis in the oldercases dealing with this common law concept had been on the cause(general intimidation) rather than on the effect (the prevention of themajority of electors from electing the candidate whom theypreferred), with time, the emphasis appears to have shifted fromcause to effect. A possible rationale for that is perhaps that persons
sc
Slrimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
125
masquerading as supporters for a particular candidate could wellhave brought about that cause, as for instance in electorates of smallsize such as boroughs with respect to which many of the earlyEnglish cases had been decided, and the consequence then wouldhave been to cause prejudice to an innocent successful candidatewho found his victory taken away by a calculated contrivanceintended to secure that objective.
In 1949 there was passed in England, The Representation of thePeople Act 1949, section 142 of which reads thus:-
"142 (1) Where on an election petition it is shown that corrupt orillegal practices or illegal payments, employments or hiringscommitted in reference to the election for the purpose ofpromoting or procuring the election of any person thereat haveso extensively prevailed that they may be reasonably supposedto have affected the result, his election if he has been electedshall be void and he shall be incapable of being elected to fillthe vacancy or any of the vacancies for which the election washeld.
An election shall not be liable to be avoided otherwisethan under this section by reason of general corruption, bribery,treating or intimidation."
The effect of subsection (2), it is to be seen, is to limit thevoidability of an election on grounds of general intimidation, toinstances where such intimidation is committed for the purpose ofpromoting or procuring the election of a particular person at suchelection. As I said, there could well have been this shift in theemphasis with the passage of time from cause to effect as respectsthe English common law concept of a free and fair election, and thenThe Representation of the People Act 1949, gave it a stamp ofstatutory authority in section 142 (1) by providing that the avoidanceof the election was to be dependent on the effect, namely theaffectation of the result of the election.
The section also introduced a guideline as to how it mayreasonably be supposed that there has been an affectation of the
126
Sri Lanka Law Reports
[1992] 2 Sri L.R.
result. The words, “corrupt practices . . . have so extensivelyprevailed that they may be reasonably supposed to have affected theresult" bring out that notion. There is therefore, an inference that iscapable of being drawn as to the affectation of the result from proofof the extensive nature that prevailed of the corrupt practice, generalintimidation being an example of such a corrupt practice.
Mr. H. L. de Silva, if I understood him correctly, was prepared toaccept that the test implicit in this section i.e. section 142 (1) of TheRepresentation of the People Act 1949, is not dissimilar to that whichconfronts the Court endeavouring to decide whether by reason ofgeneral intimidation the majority of electors may have beenprevented from electing the candidate whom they preferred, usingthe meaning given by him to these words. Though that was withrespect to the English common law concept of a free and fairelection, the same must likewise apply to section 91(a) of thePresidential Elections Act, that .section being in his submission, astatutory embodiment of that concept. If however one were toexamine the judgment in Woodward v. Sarsons one sees nowherestated, that such a test was one which a court had to adopt indeciding the question whether the majority of electors were or mayhave been prevented from electing the candidate whom theypreferred. Similarly our statutory provision namely, section 91(a) ofthe Presidential Elections Act, nowhere suggests such a test in thelanguage which the legislature thought fit to use in framing thesection. It is possible to imagine that The Representation of thePeople Act 1949 introduced this test in this statutory form in view ofthe fact that in the application of the common law concept, there wasnot implicit that test. The effect of what Mr. H. L. de Silva suggestedis, as it seems to me, to ask the Court to imply the existence of sucha test from the words appearing in section 91(a) when the actualwords of the section do not so indicate. The obvious question thatarises then is as to why, if that was the legislative intent, the sectionwas not structured so as to incorporate such a test. To use languagesimilar to that contained in section 142 (1) of The Representation ofthe People Act 1949, by what some may term a process of judicialinterpretation, would to my mind be in effect to assume the function ofthe legislature. For myself, I do not read and therefore cannot rewritesection 91(a) of the Presidential Elections Act to introduce such a test
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
127
as would enable the effect, which is that the majority of electors wereor may have been prevented from electing the candidate whom theypreferred, to be gauged by the extent of the cause, which is thedegree of prevalence of general intimidation. Had that been theobject of the legislature, the achievement of such object would havebeen a matter of simple accomplishment by the choice of languageappropriate to it, and that which the legislature has not done. I do notfeel free to do.
It was submitted by Mr. H. L. de Silva that unless the section isinterpreted in the manner suggested by him, this being the onlyprovision in the statute having the effect of sustaining the freedom offranchise (section 91(b) already having been emasculated in thisrespect by having its provisions rendered incapable of proof), aserious blow will be struck upon the democratic base of our politicalsystem. I am not convinced that this is so. It will be seen that the '•effect of section 142 (2) of The Representation of the People Act 1949is to exclude general intimidation as a ground of avoidance of anelection in England when such intimidation such as of a preventivekind, has been practised for a purpose other than the purpose ofpromoting or procuring the election of a particular person at suchelection. If one were to take the case of the petitioner as presentedtherefore, that the general intimidation complained of was calculated,not to benefit a particular candidate, but rather was of a kindintended to prevent all voters from voting, than under the law in force. in England after The Representation of the People Act 1949, therewould be no basis for avoiding this election. I find it difficult to believethat in England which is often held up as an example of a democracyat work, if it was thought that the effect of this provision was to bringabout the consequence of weakening the democratic structure oftheir government, it would have been introduced into the law.
The Indian example once again demonstrates the same thing.Mr. H. L. de Silva very properly brought to our notice the currentposition in that country. Section 100 of The Representation of thePeople Act 1951 contained a provision for declaring an election voidif it was shown that it had not been a free one by reason of thecorrupt practices of bribery or undue influence having prevailedextensively. However by The Representation of the People (Second
128
Sri Lanka Law Reports
[1992] 2 Sri L.R.
Amendment) Act 1956 this provision was removed and the positionnow in India appears to be somewhat akin in this regard to what is tobe found in England after 1949. India too is often pointed to as acountry in our part of the world where democratic institutions areseen to function, and this despite the change in the law in this regard.
Since the matter of the need to ensure the preservation ofdemocratic principles has been brought up, I think it is not out ofplace to make this comment. No doubt, the Presidential Election ofDecember 1988 was not held when conditions were ideal in thecountry. But there was a time frame within which the election had tobe held, with nothing to indicate that things would have improved inthe available short term. What then were the alternatives facing thecountry? The alternatives were, between holding the election at thistime, and not holding it before the last available date, a few weeksbeyond the date on which it was actually held. If the latter alternativewas chosen what would have been the result? Where would theenormous powers granted to the President under the Constitutionhave come to reside? Certainly not in the hands of one elected by theelectors to exercise those powers. •
•The meaning of an expression must be arrived at as a matter ofconstruction and interpretation of the language used. That should notbe done, in my understanding, by an inquiry as to how one interpretsor draws conclusions from a given set of facts and then from whatemerges, strike upon the meaning of the expression. That would, as Isee it, not be the method appropriate to interpretation. Havingdecided upon the meaning of an expression, semantically and as amatter of language, one must proceed to examine whether a givenset of facts falls within the meaning of the expression as determined.That is the method I think appropriate to statutory interpretation. Inadopting that approach, courts have often said that if a section isunworkable or leads to an absurd result, the remedy lies elsewhere.The Court has interpreted the language of the section as it falls to beread within the four corners of the statute. Even if the meaninggathered leads to an absurd result, there is nothing the Court can doabout it.
sc
Sinmavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
129
The method of interpretation contended for on behalf of thepetitioner is, as I have been able to understand, first to look at thefacts and then from the conclusions drawn therefrom, strike upon themeaning of the section. I have not been able to find in Mr. H. L. deSilva’s submissions any attempt at a clear interpretation of the words,"the majority of electors were or may have been prevented fromelecting the candidate whom they preferred”, that is, an interpretationto ascertain the meaning of the words. The closest to aninterpretation that I do find is a statement that, the words bring out thedegree of substantiality in the effect of the vitiating factors or putdifferently, that they prescribed the index or measure of the effectwhich the law requires that the vitiating factors must have on theminds of the voters. This is however not the kind of interpretation Ihave in mind, which is as to an interpretation of the language as onereads and understands it. Though undoubtedly done with admirableskill, the method he adopted was this. He said there is the commonlaw concept of a free and fair election and that the dominant attributeof that concept is the freedom of franchise. He then said that therewas at this election a shortfall in the total voter turn-out attributable togeneral intimidation, the amount of such shortfall being many timesthe winning candidate's majority and that therefore, there isdemonstrated that there was no freedom of franchise, thus making itpossible to say that the majority of electors may well have beenprevented from electing the candidate whom they preferred. How hesought to derive support for what he so contended, he did notattempt to show, as far as I could see, upon a semantic examinationof the section itself, although he said that the facts as stated by himdemonstrate that the majority of electors may have been preventedfrom electing the candidate whom they preferred. Rather, he invokedthe maxim ut res magis valeat quam pereat and stated that thesection understood in the way he contended against, results in itbeing reduced to a dead letter but understood the way he contendedfor meets the requirements of any "fact situation" contemplated by it.The maxim ut res magis valeat quam pereat in its complete formreads thus: verba ita sunt intelligenda ut res magis valeat quampereat, which means that, words are to be so understood, that theobject may be carried out and not fail. To understand "words” in thatway, the language used must show that plain meaning. The basiccanons of construction demand that the language as found in the
130
Sri Lanka Law Reports
11992) 2 Sri LR.
section must be interpreted as found. That does not mean that onecould interpret language so as to introduce words not found. As Ihave said, adopting that correct approach to interpretation may wellresult in the provision being reduced to a dead letter, but that cannotbe helped. The remedy lies in the hands of others, not in those of. theCourt. Maxwell in his work “The Interpretation of Statutes”, 12thEdition (1976) at pages 28 and 29, gives the effect of the authoritiesthus
“The first and most elementary rule of construction is that it isto be assumed that the words and phrases of technicallegislation are used in their technical meaning if they haveacquired one, and otherwise in their ordinary meaning, and thesecond is that the phrases and sentences are to be construedacqording to the rules of grammar. It is very desirable in allcases to adhere to the words of an Act of Parliament, giving tothem that sense which is their natural import in the order inwhich they are placed; The length and detail of modernlegislation1, wrote Lord Evershed M. R., ‘has undoubtedlyreinforced the claim of literal construction as the only safe rule1.If there is nothing to modify, alter or qualify the language whichthe statute contains, it must be construed in the ordinary andnatural meaning of the words and sentences. The safer andmore correct course of dealing with a question of construction isto take the words themselves and arrive if possible at theirmeaning without, in the first instance, reference to cases.
The rule of construction is to intend the Legislature to havemeant what they have actually expressed. The object of ailinterpretation is to discover the intention of Parliament, but theintention of Parliament must be deduced from the languageused,…
Where the language is plain and admits of but one meaning,the task of interpretation can hardly be sai,d to arise. Thedecision in this case1, said Lord Morris of Borth-y-Gest in arevenue case, ‘calls for a full and fair application of particularstatutory language to particular facts as found. The desirabilityor the undesirability of one conclusion as compared with
sc
Sirimavo Bandaranaike v. Ranaslnghe Premadasa and Another
(S. B. Goonewardene, J.)
131
another cannot furnish a guide in reaching a decision’. Where,by the use of clear and unequivocal language capable of onlyone meaning, anything is enacted by the legislature, it must beenforced however harsh or absurd or contrary to commonsense the result may be. The interpretation of a statute is not tobe collected from any notions which may be entertained by theCourt as to what is just and expedient: words are not to beconstrued, contrary to their meaning, as embracing orexcluding cases merely because no good reason appears whythey should not be embraced or excluded. The duty of theCourt is to expound the law as it stands, and to 'leave theremedy (if one be resolved upon), to others’.
The general rule may be stated thus: If clear in meaning, aconstruction will be adopted according to that clear meaning eventhough the ultimate result may be unjust, ab,surd or inconvenientAttorney General v. Prince Ernest Augustus of Hanover1*". If howeverwords are not plain in meaning and there are two possible alternativeinterpretations open to the Court, an interpretation which would givethis result would be avoided (Amoah Ababio v. Turkson)<42). It is onlywhen the Court has an alternative that the question of strict orbenevolent interpretation arises and not when the words are clear inmeaning. Warburton v. Loveland m.
I will now turn to certain local cases cited to ascertain how thejudges who decided them have understood the principles involved.One significant and important difference between these cases andthe present one is that in all of them the intimidation was calculated tobe beneficial to one candidate or prejudicial to another.
The first of them would be the case, of Ratnam v. DingiriBanda l5).That was a case where the charge against the respondentwas one of general intimidation as contemplated by Article 53 of the(State Council Elections) Order in-Council 1931. The petitioner aCeylon Tamil received 11093 votes and the respondent a KandyanSinhalese 12652 votes while two other Kandyan Sinhalesecandidates received 1484 and 204 votes respectively. Hearne, J.who tried the case found as a fact that the intimidation was directed
132
Sri Lanka Law Reports
[1992] 2 Sri L.R.
towards obstructing the Indian voters in the expectation that theirvotes would have been cast for the petitioner, and that that wouldhave been detrimental to the success of the candidate who wasdeclared to have been elected, namely the respondent. He held thatfreedom of choice is essential to the validity of an election and that ifby intimidation of voters that freedom was prevented generally, theelection would be void. Upon the facts before him, he found thatthere had been a deliberately planned obstruction of these voters inadvance, which had the effect of eliminating this freedom. In avoidingthe election he was concerned with, the facts as found being thatthere had been gross intimidation and that it was widespread in theareas where the petitioner had good reason to count upon heavyvoting in his favour, he concluded that it might well have preventedthe majority of the electors from returning the candidate whom theypreferred. It is seen therefore that as a matter of reasonableprobability Hearne J. was prepared, on the basis of politicalpreference, to hold in the circumstances that prevailed, that thevoters who were prevented from voting might have, if they did vote,done so for the petitioner (involving here an element of speculation),so as to prevent the returned candidate’s victory (implying that theresult was affected).
Pelpola v. Gunawardenam involved an election which was astraight contest between the respondent and the petitioner andresulted in a victory for the respondent by a comparatively narrowmargin of 387 votes. One of the allegations upon which the electionwas sought to be avoided was that of general intimidation. The officerin charge of a particular polling station, by reference to his recordsdemonstrated to the Court that out of 1427 registered voters, only 541did in fact vote. The evidence showed that out of a total of 32734voters in the whole electorate some 8375, that is over one fourth wereIndian estate labourers against whom, as a body, the acts ofintimidation in the electorate had been directed. It was alsoestablished that the Indian Congress Committee of a particular estatedecided to support the petitioner at the election and that all thelabourers had decided to vote for him. The Court also considered itnot unreasonable to suppose that the Indian labourers of theneighbouring estates which included some who were proved to havebeen molested, had likewise decided to support the petitioner. The
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
133
evidence suggested that the intimidation had an effect on thesevoters and that if 400 more persons had voted and cast their votes forthe petitioner he would have won the election. Windham, J. who wasconsidering section 77 (a) of the Ceylon (Parliamentary Elections)Order-in-Council, 1946, which is in like terms as section 91(a) of thePresidential Elections Act, in setting aside the election' concludedthus:
"To establish such a charge, where the general intimidationconsists, as here, of local acts or threats of violence, it is onlynecessary for the petitioner to show that, having regard to themajority obtained, and the strength of the polling, the result mayreasonably be supposed to have been affected. On the figuresand in the circumstances disclosed in the present case, it is atthe very least reasonable to suppose that the result of theelection may have been affected by the acts of intimidationagainst the Indian estate labourers.”
Here too, one finds that the Court was prepared, on the basis ofparty affiliation, to consider it reasonable to suppose that the voterswho were prevented from voting would have voted for the petitioner,once again involving an element of speculation. This election too wasset aside on the basis that the result of the election may reasonablybe supposed to have been affected, meaning that the reasonableprobabilities were that the returned candidate would not have beenentitled to success.
The case of Tarnolis Appuhamy v. Wilmot Perera 171, which I havealready referred to, was one pertaining to an election where threecandidates contested the seat in question. The allegation there hadbeen that there was general intimidation in terms of section 77(a) ofthe Ceylon (Parliamentary Elections) Order in Council 1946.Nagalingam, J. in trying the case concluded with respect to the facts,that no evidence was given of what may be termed coerciveintimidation, that is intimidation having for its object the use of forceor threat to compel a voter to vote for a particular candidate, but whatevidence was led was led to show that the election was subjected topreventive intimidation, that is intimidation which had for its object theprevention of electors from going to the poll lest the rival candidate
134
Sri Lanka Law Reports
[1992] 2 Sri L.R.
should get their votes, and that having regard to the number polledand to the circumstance that this electoral area had annexed to itselfthe highest percentage of voters in any electoral area in the island itwas difficult to convince anyone that voters in general were deterredby anything savouring of intimidation from going to the polls orrecording their votes. Nagalingam, J. did make this comment, one inpoint of fact unnecessary for the purposes of his decision:- “I mustnot, however, be understood as saying that if it is shown that, thougha large number may have polled nevertheless a fair number of theelectorate were prevented from exercising their right freely, that wouldnot by itself be a sufficient ground for declaring the election void, butof this there is scarcely any proof in this case." Yet in refusing toavoid the election he cited with approval, the statement of Gibson, J.in the North Louth Case {supra) that “to upset an election for generalintimidation it is necessary to show that there was such generalintimidation as might have affected the result of the election".
Of much importance in this regard is also be the case ofllahgaratne v. De Silva(2). There was an allegation in that case that byreason of circumstances attending on or following the recent floods inthe District, including the disorganisation of the life of large sectionsof the voters, the segregation of refugees who were voters,
■ disturbance of communication and transport and the scarcity ofpetrol, the majority of the electors were or may have been preventedfrom electing the candidate whom they preferred. Section 77(a) of theCeylon (Parliamentary Elections) Order-in-Council 1946 was onceagain invoked by the petitioner to have the election set aside.Windham, J. was influenced by the fact of there being no evidencefrom which to ascertain whether the inmates of the refugee campsand also those other homeless persons who found refuge with friendsdid in fact refrain from going to the poll and if so why; no evidence toshow what number of these persons and of the other refugees or ofother persons affected by the floods were voters; and no evidence toshow how many of such of them as were voters abstained frompolling or that if any did so abstain it was by reason of circumstancesarising from the floods. After analysing the evidence pertaining to thischarge he concluded (at page 184) thus:-
“ Accordingly I cannot hold on the evidence that the majorityof the electorate were or may have been prevented from
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
135
electing the candidate they preferred by reason of thecircumstances having prevented them from voting for anycandidate at all”, (my emphasis)
These words, in particular those emphasized by me, have to mymind a high degree of relevance to the instant case, where it isclaimed that the general intimidation was directed to prevent votersfrom voting at the election and thus to prevent them from voting forany candidate at all.
In Jayasinghe v. Jayakodyw Sharvananda, C.J. (at page 89) said:
“If it is proved that a corrupt practice had been committed bythe returned candidate or an election agent or by any otherperson with the knowledge or consent of the returnedcandidate, then the Election Judge has to declare the electionvoid. But if the corrupt practice had been committed by a.person other than the persons mentioned in section 77(c) (ofthe Ceylon Parliamentary Elections Order-in-Council 1946) thenit must be further established that the majority of electorsthereby were or may have been prevented from electing thecandidate whom they preferred, for the Election Judge todeclare the election void.”
Thus I come to the conclusion that the words in section 91(a) ofthe Presidential Elections Act which are, “the majority of electors wereor may have been prevented from electing the candidate whom theypreferred," have the meaning that, had the constituency had a freeand fair opportunity of electing the candidate the majority preferred,that candidate would or may have been someone other than thereturned candidate. This would then mean that this expression thatoccurs in section 91(a), has the same meaning as the expression,“affected the result of the election” as those words appear in section91(b) of the Presidential Elections Act, and that the order of the Courton the preliminary objections intended to say so. The point ofdifference is that while the words in section 91(b) have reference to asituation of certainty, those in section 91(a) while equally havingreference to a situation of certainty, also deal with a situation ofprobability, which I have interpreted as reasonable probability.
136
Sri Lanka Law Reports
[1992] 2 Sri L.R.
I have already considered the case of the petitioner on the basis ofthe meaning contended for her with respect to the words, “themajority of electors were or may have been prevented from electingthe candidate whom they preferred" and concluded that she hasfailed to discharge the burden placed on her upon that construction.
The position that then arises upon the meaning I ascribe to thisexpression is that a burden is cast on the petitioner under section91(a) to show that but for the factor or factors complained of, beinggeneral intimidation and non-compliance with the provisions of theelections law, or even a combination of them, the successfulcandidate would or may reasonably have been someone other thanthe 1st respondent.
Then the next question that may be thought to arise is how onecould prove this essential ingredient when the provisions of section91(a) are invoked. To answer a question of that kind, I cannot see asbeing the function of a court. The Court interprets the legal provisionas found and then having decided upon its meaning, embarks uponan examination of the facts, as found to see whether the fact to beproved, has in fact been established. If not proved in the Court’sview, the case set up must fail. If the factum probandum is incapableof proof, the remedy lies elsewhere. The position is explained byGhulam Hasan, J. of the Indian Supreme Court in the case of VashitNarain Sharma v. Dev Chandra malthough the case was withrespect to a question of an improper acceptance of a nominationpaper, thus
“The question is one of fact and has to be proved by positiveevidence. If the petitioner is unable to adduce evidence in acase such as the present, the only inescapable conclusion towhich the Tribunal can come is that the burden is notdischarged and that the election must stand. Such result mayoperate harshly upon the petitioner seeking to set aside theelection on the ground of improper acceptance of a nominationpaper, but neither the Tribunal, nor the Court is concerned withthe inconvenience resulting from the operation of the law.
How this state of things can be remedied is a matter entirely forthe legislature to consider".
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
137
The petitioner’s position appears to be, as I have been able tounderstand it, that there is no evidence before the Court to show anyaffectation of the result as a consequence of general intimidation. Ifher own assessment of the evidence placed by her before the Courtis to that effect as is seen by the fact that she has not sought toidentify any relevant items of evidence in support, it would be adifficult task indeed for others to undertake a search for suchmaterial.
Though a statement to that effect might have been sufficient in theordinary kind of case to draw a conclusion unfavourable to thepetitioner, and thus to dispose of the matter, in a case such as thisthere is a duty cast on the Court which it should not refrain fromperforming, and that is to examine the evidence though notnecessarily in detail to see what relevant conclusions could be drawnas to this aspect. There is no clear evidence which does furnishdirect proof of this requirement that I can see, and so I must askmyself whether there is any reasonable inference that could be drawnfrom the evidence considered as a whole as to an affectation of theresult.
Before addressing my mind to that question it is necessary toconsider a submission made by Mr. H. L. de Silva which I havealready referred to that any exercise to determine whether theelement that the result of the election may have been affected, in thesense that but for the acts complained of another candidate otherthan the one declared returned would have been successful, wouldinvolve a computation of numbers which necessarily must be basedupon conjecture and surmise, being dependent upon circumstancestotally different and unpredictable, and therefore not something thatsection 91(a) of the Presidential Elections Act could reasonably bethought to demand. He placed much emphasis upon certainpassages (at pages 21-23) from the judgment in the case of ShivCharan Singh v. Chandra Bhan Singh (,2> decided by the SupremeCourt of India. They read thus
“ The burden to prove this material effect is difficult and manytimes it is almost impossible to produce the requisite proof. Butthe difficulty in proving this fact does not alter the position oflaw. The legislative intent is clear that unless the burden
138
Sri Lanka Law Reports
[1992] 2 Sri L.R.
howsoever difficult it may be, is discharged, the election cannotbe declared void. The difficulty of proving the material effectwas expressly noted by this Court in Vashist Naraian Sharmaand Paokai Haokip cases and the Court observed that thedifficulty could be resolved by the legislature and not by thecourts. Since then the Act has been amended several times, butParliament has not altered the burden of proof placed on theelection petitioner under section 100 (1) (d) of the Act. Thereforethe law laid in the aforesaid decisions still holds the field. It isnot permissible in law to avoid the election of the returnedcandidate on speculations or conjectures relating to the mannerin which the wasted votes would have been distributed amongthe remaining validly nominated candidates. Legislative intent isapparent that the harsh and difficult burden of proving materialeffect on the result of the election has to be discharged by theperson challenging the election and the courts cannotspeculate on the question. In the absence of positive proof ofmaterial effect on the result of the election of the returnedcandidate, the election must be allowed to stand and the Courtshould not interfere with the election on speculation andconjectures … it is difficult to comprehend or predicate withany amount of reasonable certainty the manner and theproportion in which the voters who exercised their choice infavour of the improperly nominated candidate would haveexercised their votes. The courts are ill-equipped to speculateas to how the voters could have exercised their right of vote inthe absence of the improperly nominated candidate. Anyspeculation made by the Court in this respect would bearbitrary and contrary to the democratic principles. It is a matterof common knowledge that electors exercise their right of voteon various unpredictable considerations. Many times electorscast their vote on consideration of friendship, party affiliation,local affiliation, caste, religion, personal relationship and many ■other imponderable considerations. Casting of votes by electorsdepends upon several factors and it is not possible to forecastor guess as to how and in what manner the voters would haveexercised their choice in the absence of the improperlynominated candidate.”
Although the greater part of this quotation is suggestive of mattersfavourable to the 1st respondent insofar as they help to sustain an
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
139
election, it is the latter part that Mr. H. L. de Silva really called in aid.When he made reference to these passages, I believe they also hada contextual relevance to his submission that arising from the shortfallin the voter turn-out consequent upon general intimidation whencompared with the majority of votes which the 1st respondentsecured, an inference could be drawn adverse to the latter, that themajority of electors might conceivably have chosen differently. Hecontended if I mistake not, that such an inference could not berebutted by looking at past performance of voters at earlier electionsand that voting patterns at subsequent elections are all unsafe andunreliable as a guide.
The circumstance with reference to which the Indian SupremeCourt came to use the words it did contain in the latter part of thispassage, which as I said was what Mr. H. L. de Silva relied onespecially, is indicated in the judgment thus
“It is manifest that law laid down by this Court in VashitNarain Sharma case and Paokai Haokip case holds the fieldand it is not permissible to set aside the election of a returnedcandidate under section 100 (1) (d) on mere surmise andconjectures”.
Two things therefore are apparent from this passage, firstly andprimarily, that the Court was giving here a justification for declining toset aside the election, in other words stating its refusal to take theserious step of setting aside an election upon the basis of materialessentially speculative in nature and, secondly, that the Court wasfollowing the decisions in two earlier cases, which I myself will nowrefer to.
In the case of Vashit Narain Sharma v. Dev Chandra m which Ihave already referred to the Court was concerned with theinterpretation of a provision which enabled an election to be set asideupon proof that “the result of the election has been materiallyaffected", in a context where a candidate's nomination had beenimproperly accepted and consequently the votes that he received atthe election had been wasted. The Court (at page 515) said:
140
Sri Lanka Law Reports
[1992] 2 Sri L.R.
"These words seem to us to indicate that the result should notbe judged by the mere increase or decrease in the total numberof votes secured by the returned candidate but by proof of thefact that the wasted votes would have been distributed in sucha manner between the contesting candidates as would havebrought about the defeat of the returned candidate".
In my understanding of the judgment, these words constitute theCourt’s statement as to the mode of proof that “the result of theelection has been materially affected”. It is useful to note that theCourt added (at pages 515 & 516) thus
“But we are not prepared to hold that the mere fact that thewasted votes are greater than the margin of votes between thereturned candidate and the candidate securing the next highestnumber of votes must lead to the necessary inference that theresult of the election has been materially affected. That is amatter which has to be proved and the onus of proving it liesupon the petitioner. It will not do merely to say that all or amajority of the wasted votes might have gone to the nexthighest candidate.
The casting of votes at an election depends upon a variety offactors and it is not possible for anyone to predicate how manyor which proportion of the votes will go to one or the other of thecandidates. While it must be recognised that the petitioner insuch a case is confronted with a difficult situation, it is notpossible to relieve him of the duty imposed upon him by section100 (1) (c) and hold, without evidence that the duty has beendischarged. Should the petitioner fail to adduce satisfactoryevidence to enable the Court to find in his favour on this point,the inevitable result would be that the Tribunal would notinterfere in his favour and would allow the election to stand".
The case of Paokai Haokip v. Rishang m was also concernedwith the interpretation of these identical words “the result of theelection has been materially affected” occurring in the samelegislative enactment. The grounds upon which the electionwas attacked were, that the polling had been disturbed due to
sc
Sirimavo Bandaranalke v. Ranasinghe Premadasa and Another
(S. B. Goonewardene. J.)
141
numerous circumstances such as shutting of polling stations withoutdue notification thereof, and that there had been disruption of votingat certain polling stations due to firing of guns. In deciding the case,the Court considered the question whether the burden of proof hadbeen discharged by the petitioner by demonstrating to the Courteither “positively” or even “reasonably”, that the poll would have goneagainst the returned candidate but for the matters complained of. Insearching for an answer to the test so adopted, the Court looked atcertain matters the nature of which the following words in thejudgment (at page 666) show:
“To begin with, it is wrong for the election petitioner tocontend that of the 6,726 votes which were not cast, he wouldhave received all of them. The general pattern of polling notonly in this constituency but in the whole of India is that all thevoters do hot always go to the polls. In fact, in this case, out of219,554 voters only 120,008 cast their votes. Even if we were to-add to them the 6,726 votes, it is obvious that not more than55% of the voters would have gone to polls. This immediatelycuts down the figure of 6,726 to a little over half and the marginfrom which the election petitioner could claim additional votestherefore becomes exceedingly small. When we turn to thepattern of voting as is disclosed in the various polling booths atwhich the voters had in fact gone, we get a reasonably clearpicture. At 9 polling centres, 1893 votes are actually polled. Ofthese, 524 votes were received by the election petitioner and413 by the returned candidate and 1097 votes went to the othercandidates. In other words, out of 20 votes 11 went to othercandidates, 5 to the election petitioner and 4 to the returnedcandidate. If one goes by the law of averages and appliesthese figures reasonably to half of the votes which were notcast, it is demonstrated at once that the election petitioner couldnot expect to wipe off the large arrears under which he labouredand that he could not have therefore made a successful bid forthe seat even with the assistance of the voters who did not casttheir votes”.
Of importance is also the following passage at page 667 whichreads thus
142
Sri Lanka Law Reports
[1992] 2 Sri LR.
“That section requires that the election petitioner must go alittle further and prove that the result of the election had beenmaterially affected. How he has to prove it has already beenstated by this Court and applying that test, we find that he hassignificantly failed in his attempt and therefore the election ofthe returned candidate could not be avoided. It is no doubt truethat the burden which is placed by law is very strict; even if it isstrict it is for the courts to apply it. It is for the Legislature toconsider whether it should be altered. If there is another way ofdetermining the burden, the law should say it and not thecourts. It is only in given instances that, taking the law as it is,the courts can reach the conclusion whether the burden of proofhas been successfully discharged by the election petitioner ornot”.
. I think these passages, effectively demonstrate that the Court hadbeen disinclined to set aside elections, even when upon itsinterpretation of the legislative provision concerned, a virtuallyimpossible burden had been cast upon the petitioner, but in its effortto search for the relevant material, the Court had not hesitated toconsider figures, statistics, averages etc. all of which involve ameasure of speculation and surmise.
In the cases of Ratnam v. Dingiri Banda 151 and Pelpola v.Gunawardene(B) the Court was prepared to consider it reasonable tosuppose that the voters who were prevented from voting would ormay have voted for the unsuccessful candidate had it not been forthe acts complained of, once again involving an element ofspeculation.
In the North Louth case I have already referred to, Mr. Justice' Madden thought it proper to take into consideration voting figuresrelating to other elections.
The petitioner’s own position was that having regard to the nationalaverage of voting figures demonstrated in the post-independentperiod in this country, it is reasonable to state that one could haveexpected 80% of voters from and out of the total number ofregistered voters to have voted. When one endeavours in this manner
sc
Sirimavo Bandaranalke v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
143
to fix upon a particular figure such as 80%, is there not involved in itsomething in the nature of a statistical exercise?
There may be something to be said for any criticism of thestatistical method, because conclusions based upon statistics, notonly in this area but in other areas as well, have been demonstratedtime and again to be erroneous. Indeed in the case of Paokai v.Rishangm already referred to Hidayatullah, C.J. (at page 666) usedthese words of caution
"While we do not think that statistics can be called in aid toprove such facts, because it is notorious that statistics canprove anything and made to lie for either case, it is open to us inreaching our conclusion to pay attention to the demonstratedpattern of voting".
I believe that in a case such as this where one attempts to drawinferences, one inevitably runs into an uncertain area of conjecture,surmise and speculation but that is something I feel that has to beaccepted as inevitable. As I have already pointed out, statistics mustalways be viewed with caution when made the basis of conclusions.Yet, in dealing with what are necessarily hypothetical situations, suchas where one endeavours to glean as to what the position might havebeen if things had been different, some use of statistics I think ispermissible and indeed unavoidable. We have to look at them andsee whether they give us any guidance in this difficult area as towhether the election should be avoided, because if one were tototally disregard them, there may then be no guiding materialwhatever.
Before commencing to look at the evidence, it may perhaps beuseful to state something as to what the approach on behalf of the1st respondent was. It was claimed for him that there is nojustification whatever for avoiding the election, a task which a Courtwill not in any event lightly undertake. The endeavour was to showthat any reasonable interpretation of the evidence would produce theconverse result. Mr. Choksy sought to point out that the brunt of theintimidation was directed at the 1st respondent’s campaign, hissupporters and his party and that the adverse effects thereof were by
144
Sri Lanka Law Reports
[1992] 2 Sri L.R.
far upon him, and that these factors taken in conjunction with whatthe voting patterns indicate, must result in the petitioner's case failing.Indeed it was the 1st respondent’s position that the acts ofintimidation and threats of intimidation including intimidatory posterswhich were directed at the United National Party and its supporters,commenced at an earlier point of time, even prior to the ProvincialCouncils elections which the petitioner's party did not contest, andthat its momentum continued throughout the period of thePresidential Election of 1988 so that it was the members andsupporters of the 1st respondent’s party who during this PresidentialElection had every reason to be genuinely intimidated, having regardto the situation they had to face and the experiences they underwentearlier.
Mr. Choksy contended that if one has regard to the votingpercentages in the 1989 general election two things aredemonstrated, firstly, that the increased percentages indicate thatthere was an improvement in the conditions prevailing in the countryand, secondly, that such improvement which resulted in a greatervoter turn-out was beneficial to the United National Party. He referredto the evidence of certain witnesses who testified on the lines thatthere was such an improvement, evidence which I do not thinknecessary to reproduce here. He also contended that there was thisunusual feature in this case which does not find a parallel anywhereelse, that the 1989 general election was held about three monthsafter the Presidential Election of 1988 and that both elections wereheld on the same electoral register and with respect to the sameelectoral area, namely, the whole island. He submitted that theconclusion that has to be drawn is inevitable, that with the increasedvoter turn-out there was a remarkable improvement in theperformance of the United National Party in the 1989 generalelection, as contrasted with the 1st respondent’s performance at thePresidential Election of 1988, which itself showed, having regard tothe brief time interval between these two elections, that whatevershortfall there was in the voting at the 1988 Presidential Election, itworked to the disadvantage of the 1st respondent. The point hesought to make was that the volume of evidence which he ledregarding intimidation, showed it to be directed against the UnitedNational Party, its supporters, its trade union and other organisations,
sc
Sirimavo Bandaranaike v. RanasSnghe Premadasa and Another
(S. B. Goonewardene, J.)
145
its members and those connected with it, all of which had thisadverse effect on the 1st respondent's performance at thePresidential Election and that corroboration of that position was seenby the lower voter turn-out in 1988 which gave the 1st respondentpercentage of the votes cast, which was considerably less than thatwhich the United National Party received a mere three months later atthe general election, when conditions as he claimed were better.
It must not be lost sight of that the petitioner and the 1strespondent, though they contested the election as individuals, werein reality representatives put forward by two major political parties inthe island and therefore their candidature must be viewed in thatlight, so that when, one considers these figures, one must look atthem largely as an index to party preference. That being so, one mustonce again refrain from considering any previous elections wherethere were no-contest pacts operating so that there would then be noblurring of the figures.
The areas of electoral boundaries have changed from time to timewith delimitations, and thus a safe and adequate way of looking atany voting pattern would, to my mind, be by reference to Provinceswhose boundaries have undergone no changes, and when we get asfar as voting patterns in particular areas, it becomes unnecessary inmy view, to refer even generally to the evidence in the case. To beable to reach the necessary conclusions in the case in this manner,eliminates a serious disadvantage which I would otherwise havebeen labouring under, namely, not having seen or heard a goodmany of the witnesses who testified. I
I think it is apt to state once again at this point that the secrecy ofvoting provisions do not prevent the drawing of any conclusions fromthese voting patterns. It could not have been of much use to haveasked any voter as to the manner in which he did vote and that iswhere the prohibition is. If at all, it is the voter who did not vote whomight have been asked how he would have voted, if he had had theopportunity to do so, something I do not consider prohibited. Theaffectation of the result if any is as a consequence of preventingthose who did not vote from doing so, as I have pointed outelsewhere.
146
Sri Lanka Law Reports
[1992] 2 Sri L.R.
The question then is whether the voters who did not vote if they didvote, would have changed the result.
One possible index would be to look at the figures of those whohad in fact voted and say that if those who did not vote did in factvote, such extra votes could not on the evidence be expected tohave been of a greater percentage for the petitioner than that whichin the event she did receive. To say differently would be to say thatthe evidence demonstrates that those who kept away from votingwere for the most part those who would have voted for the petitioner. Ihave stated elsewhere what the implications of saying so would be, interms of the percentages that she would have had to obtain out ofthose extra votes. I do not think that the evidence in the casesuggests as reasonable such a probability. It might perhaps havebeen possible to contemplate it as reasonably probable, had therelevant evidence in the case not shown that the target of intimidationwas by far the 1st respondent’s party and his supporters. There ishowever something else that needs to be added here. A suggestionwas made in cross-examination to certain witnesses called for the 1strespondent, the object of which appeared to be to show that theUnited National Party had become unpopular by reason of its policiesand practices and the adoption of the Indo-Sri Lanka Accord, andwhat I believe was implied thereby was, that had the voters who didnot vote in fact voted, that would have been advantageous to the SriLanka Freedom Party. I am not convinced that such a point issustainable. One could not take as a reliable measure of popularityanything more than what the actual percentages of voters who didvote do suggest, and that is that the majority supported the 1strespondent. Upon that measure of popularity then as to what issuggested by the figures of the voters who did actually vote, thepetitioner clearly not having received the greater percentage of votescannot be thought to have been successful in sustaining thiscontention.
The other possible index may be that suggested by votingpatterns, in examining which I will refer to voter percentages by.reference to the Provinces. The figures I reproduce exclude theNorthern and Eastern Provinces, the complete figures of which I donot find available, but their absence makes no significant difference.
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
147
In the Western Province at the general elections in 1977 the UnitedNational Party received 55.28% while the Sri Lanka Freedom Partyreceived 31.63% of the valid votes cast. In 1982 at the PresidentialElection the United National Party received 54.07% and the Sri LankaFreedom Party received 40.76% of the valid votes cast. At the 1988Presidential Election the 1st respondent received 48.24% and thepetitioner received 47.03% of the valid votes cast. At the 1989general elections the United National Party received 52.34% and theSri Lanka Freedom Party received 35.89% of the valid votes cast.
In the Central Province at the general elections in 1977 the UnitedNational Party received 56.07% and the Sri Lanka Freedom Partyreceived 34.15% of the valid votes cast. In 1982 at the PresidentialElection the United National Party received 60.16% and the Sri LankaFreedom Party received 36.03% of the valid votes cast. At the 1988Presidential Election the 1st respondent received 57.12% and thepetitioner received 41.27% of the valid votes cast. At the 1989general elections the United National Party received 62.61% and theSri Lanka Freedom Party received 30.83% of the valid votes cast.
In the Southern Province at the general elections in 1977 theUnited National Party received 54.81% and the Sri Lanka FreedomParty received 28.88% of the valid votes cast. In 1982 at thePresidential Election the United National Party received 49.01% andthe Sri Lanka Freedom Party received 42.20% of the valid votes cast.At the 1988 Presidential Election the 1st respondent received 45.13%and the petitioner received 52.35% of the valid votes cast. At the1989 general elections the United National Party received 51.95%while the Sri Lanka Freedom Party received 40.41% of the valid votescast.
In the North Western Province at the general elections in 1977 theUnited National Party received 56.50% while the Sri Lanka FreedomParty received 35.44% of the valid votes cast. In 1982 at thePresidential Election the United National Party received 56.64% whilethe Sri Lanka Freedom Party received 39.20% of the valid votes cast.At the 1988 Presidential Election the 1st respondent received 52.87%while the petitioner received 45.21% of the valid votes cast. At the1989 general election the United National Party received 59.71%
148
Sri Lanka Law Reports
[1992] 2 Sri LR.
while the Sri Lanka Freedom Party received 35.14% of the valid votescast.
In the North Central Province at the general elections in 1977 theUnited National Party received 54.63% while the Sri Lanka FreedomParty received 39.71% of the valid votes cast. In 1982 at thePresidential Election the United National Party received 51.82% whilethe Sri Lanka Freedom Party received 40.98% of the valid votes cast.At the 1988 Presidential Election the 1st respondent received 46.26%while the petitioner received 51.80% of the valid votes cast. At the1989 general elections the United National Party received 58.16%while the Sri Lanka Freedom Party received 37.25% of the valid votescast.
In the Uva Province at the general elections in 1977 the UnitedNational Party received 58.47% while the Sri Lanka Freedom Partyreceived 37.44% of the valid votes cast. In 1982 at the PresidentialElection the United National Party received 55.87% while the SriLanka Freedom Party received 38.51 % of the valid votes cast. At the
Presidential Election the 1st respondent received 60.60% whilethe petitioner received 36.83% of the valid votes cast. At the 1989general election the United National Party received 57.14% while theSri Lanka Freedom Party received 37.69% of the valid votes cast.
In the Sabaragamuwa Province at the general elections in 1977the United National Party received 54.01% while the Sri LankaFreedom Party received 29.88% of the valid votes cast. In 1982 at thePresidential Election the United National Party received 53.95% whilethe Sri Lanka Freedom Party received 40.54% of the valid votes cast.At the 1988 Presidential Election the 1st respondent received 54.21%while the petitioner received 43.39% of the valid votes cast. At the
general election the United National Party received 59.36%while the Sri Lanka Freedom Party received 32.68% of the valid votescast.
A synopsis of the percentages which the petitioner and her partyobtained in the aforesaid elections of 1977, 1982 and 1989respectively are thus: Western Province 31.63%, 40.76%, 35.89%;
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
149
Central Province 34.15%, 36.03%, 30.83%; Southern Province28.88%, 42.20%, 40.41%; North Western Province 35.44%, 39.20%,35.14%; North Central Province 39.71%, 40.98%, 37.25%; UvaProvince 37.44%, 38.51%, 37.69%; Sabaragamuwa Province29.88%, 40.54%, 32.68%. At the Presidential Election of 1988 thepetitioner received 44.95% of the total valid votes cast. It is thereforeto be seen that her performance at that election was an improvementwhen compared with the percentages I have indicated, none of themhaving ever exceeded 44.95%. Upon the earlier basis which Iconsidered, I expressed the view that upon the evidence it would notbe reasonable to think that the petitioner could be expected to havereceived by way of a percentage out of the votes not cast, anything inexcess of the percentage she actually did receive out of the votes infact cast. Allowing her as a maximum such a percentage out of thevotes not cast is then seen to be advantageous to her, as comparedwith a basis made to depend on figures available from otherelections.
Any objective examination of these figures, subject no doubt to theinfirmities that any such exercise involves, does thereforedemonstrate one thing, which is, that it is not reasonably possible tosay that if more voters did vote in the 1988 Presidential Election, therewould have been a benefit accruing to the petitioner either by way ofsuccess at the election or indeed even by way of an increasedpercentage so that the difference in votes between those of the 1strespondent and herself would have become narrower. In that state ofthings therefore, far from saying that there was an affectation of theresult in the sense that such result would have been favourable to thepetitioner, the reasonable probabilities suggested are the reverse andthat is that a greater voter turn-out could well have benefited the 1strespondent in the sense that his majority may well have become ■greater and therefore I think the point made by Mr. Choksy is notwithout some foundation. Therefore the conclusion that I am driven todraw is that it cannot be said, even upon a review of the evidencethat the reasonable probabilities are that the majority of electors wereor may have been prevented from electing the candidate whom theypreferred in the sense of the success of another candidate other thanthe 1st respondent, that being the petitioner herself in thecircumstances of this case.
150
Sri Lanka Law Reports
[1992} 2 Sri LR.
The petitioner’s case therefore as pleaded in paragraph 6A of herpetition, upon the basis of general intimidation as a vitiating factorunder section 91(a) of the Presidential Elections Act must fail byreference to a correct interpretation of the meaning of the words “themajority of electors were or may have been prevented from electingthe candidate whom they preferred” understood as an affectation ofthe result, viewed in the light of what is suggested by the evidence.
Some of the aspects of the case that I will hereafter proceed toadvert to, would not ordinarily have required examination, but in viewof the public interest element involved, I will make some reference tothem albeit in brief.
The next aspect of the petitioner’s case is that contained inparagraph 6B of her petition which is a complaint of non-compliancewith the elections law. This, as I earlier remarked, was the ground thatwas abandoned by Mr. H. L. de Silva on the basis that the Courtinterpreted section 91(b) so as to say that affectation of the result wasnecessary to be established, and in respect of which the petitionerhas not been able to furnish the requisite proof. The Court in its orderupon the preliminary objections did hold that there was a requirementunder that section for the petitioner to establish that the result of theelection was affected by reason of non-compliance with the electionslaw. There was no claim of any need to clarify anything in this regardand therefore there arises no need to examine afresh whether it isindeed an essential requirement under section 91(b) to establishaffectation of the result. It would suffice I think to state two things,firstly, that apart from the authorities, the plain language of the sectionsupports the view of the need to establish affectation of the result andtherefore the Court’s interpretation of section 91(b) of the Act upon itsorder on the preliminary objections is correct and secondly, that thepetitioner’s own estimate of the evidence, led in the case, that there isno evidence showing affectation of the result, can undoubtedly beaccepted without debate and that in any case an independentexamination of that evidence does reveal that to be the correctposition. In the circumstances, the ground pleaded in paragraph 6Bof the petition has not been established and must fail.
The next matter to consider is the ground as pleaded in paragraph6C of the petition which is that the pleaded items of non-compliance
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
151
with the elections law also constituted a basis of avoidance undersection 91(a), as another "circumstance". In dealing with thepetitioner's case based upon general intimidation, I have also dealtwith such aspect of her case as pertains to what is pleaded inparagraph 6C of the petition and having regard to the conclusion Ihave reached with respect to'the case based upon generalintimidation, the latter question does not I think require any furtherindependent examination. The requirement being that it is incumbentunder section 91(a) to establish that the majority of electors were ormay have been prevented from electing the candidate whom theypreferred, it was incumbent on the petitioner to show that by reasonof non-compliance with the elections law referred to, theconsequence was that the wrong candidate was declared elected ormay be reasonably thought to have been declared elected. There isno evidence affording proof as to that, and therefore the case of thepetitioner as pleaded in paragraph 6C of her petition must also fail.
There then is the position taken by Mr. H. L. de Silva for thepetitioner different from what is pleaded in paragraph 6C of thepetition, that the non-compliance with the elections law pleaded wasnot something for which the 2nd respondent was accountable orresponsible, but that the acts of general intimidation complained ofresulted in a breakdown of the machinery of election, so that taken inconjunction with the general intimidation itself which led to thatbreakdown there was demonstrated another circumstance as that. word is used in section 91(a) of the Presidential Elections Act andconstituting a basis of avoidance thereunder. The learned Attorney-General Mr. Marapana, then the Solicitor-General, quiteunderstandably reacted in protest so as to indicate that that was notthe case he was called upon to meet upon the petition, and that itwas one of a radically different nature from what he had throughoutaddressed his mind to. Mr. H. L. de Silva’s response was that with thepressure brought about by the limited time period allowed by law forthe filing of the petition, the petitioner was constrained to structureher petition as best as could be done in the circumstances, but thatthe Court should nevertheless investigate this complaint as all therelevant material for doing so is before it. I am far from agreeing withwhat I may term that explanation, based upon a claim of lack of time.The petitioner, possessed as she obviously was of the material
152
Sri Lanka Law Reports
[1992] 2 Sri LR.
pleaded as constituting instances of non-compliance with theelections law, did not need much more by way of other material ortime, in order to aver that such non-compliance was the result of thegeneral intimidation complained of, without stating her case the wayit has been done on the petition.
The relevant paragraph in her petition numbered 6C which I havealready reproduced uses the words “the failure of the Commissioneror Elections (the Second Respondent) and/or certain members of hisstaff to conduct a fair and free election in accordance with thePresidential Elections Act No. 15 of 1981, more particularly set out inparagraph 9 read with paragraph 8 hereof". The plea relating to non-compliance with the elections law as a ground of avoidance undersection 91(b) of the Presidential Elections Act is what is referred to inparagraph 8 of the petition and is the abandoned ground. Paragraph8 also uses the words “by reason of non-compliance with theundermentioned provisions of the Act by the Election Commissionerand/or members of his staff and officers employed for the conduct ofthe election", suggesting that blame therefore should be laid at thefeet of the 2nd respondent. The plea relating to non-compliance withthe elections law as a basis for avoidance of the election undersection 91(a) of the Act is what is contained in paragraph 9 of suchpetition. It mentions nothing even remotely suggesting a link withgeneral intimidation.
To allow the petitioner therefore to urge a case founded upon anentirely different basis which seeks to attribute non-compliance withthe elections law to the existence of general intimidation, and thus toseek to have the election declared void, would be, as I see it totallyunfair and which I for one, do not consider as meeting therequirements of justice. Non-compliance with the elections law asanother “circumstance" as contemplated by section 91(a) of thePresidential Elections Act is one thing, but non-compliance occurringas a result of general intimidation committed by others and therebyforming an aspect of the case based upon general intimidation isaltogether another, and indeed one upon which the 2nd respondentneed not, as of necessity, have been made a party to the petition.Properly speaking the new position taken could not be characterisedas non-compliance with the elections law, but rather, would indicate
sc
Sirimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
1S3
that the 2nd respondent was rendered incapable of complying withsuch law, or that he failed to comply with such law by reason ofcauses beyond his control and therefore all the pleadings on this limbof the petitioner’s case as they appear in the petition must fail. As Ihave pointed out, the ground of avoidance under section 91(a) is oneupon which the majority of electors were or may have beenprevented from electing the candidate whom they preferred. Theagency by which or the cause as a result of which the majority ofelectors were or may have been so prevented could be theprevalence of general intimidation or a non-compliance with theelections law. If it is the petitioner's case that such agency or causewas general intimidation, which in turn brought about a failure of themachinery of election, which once again produced the effect that themajority of electors were or may have been prevented from electingthe candidate whom they preferred, the petition as a matter ofpleadings should have contained a clear statement on those lines.What the petitioner endeavours therefore is, to do indirectly what shewould not have been permitted to do directly, that is to present acase different from that pleaded in her petition, and so in effect,amending such petition in this very important regard. In principle, topermit such a course would not be to conform either to the letter or tothe spirit of the law. Nevertheless it may usefully be added that evenif the petitioner’s case be looked at in this way, she does notovercome the effect of the absence of evidence to show that themajority of the electors were or may have been prevented fromelecting the candidate whom they preferred, the factual position herebeing that which I pointed out with regard to her case based ongeneral intimidation.
As stated earlier, when Mr. H. L. de Silva was addressing the Courtwith respect to the petitioner’s case as contained in paragraph 6C ofthe petition based upon a claim of non-compliance with theprovisions of the elections law as another circumstance upon which itwas sought to have the election set aside under section 91(a), headopted a new approach to the case, subject to one claimedexception regarding which as I have said there was no clearstatement nor any submissions made, so that there is no strict needto deal with that exception. However for the sake of completeness,something may be said about what I was able to gather to be that
154
Sri Lanka Law Reports
[1992] 2 Sri LR.
exception, which was a reference to the electoral district ofMoneragala. The allegation in the petition is that the polling in 49polling stations was declared null and void by the 2nd respondentresulting in a failure to comply with section 46A of the Act or thatthere was a failure to take into account that the votes polled at thosepolling stations or those which would have been polled, would haveaffected the result of the election. It was accepted by Mr. Marapana.that in the circumstances prevailing, the 2nd respondent wascompelled to do so, acting under the provisions of section 46Aintroduced by way of amendment into the Presidential Elections ActNo. 15 of 1981 by the Elections (Special Provisions) Act No. 35 of1988. Section 46A makes provision for a situation as prevailed atthose polling stations and it empowers the Commissioner of Electionsto act as he did, provided that he is of the opinion that the result ofthe election for such electoral district would not be affected by thefailure to count the votes polled or the votes that would have beenpolled at such polling stations. Before doing so however, theCommissioner of Elections is called upon to consult the candidatesor their agents. In the course of evidence led for the 2nd respondent,the document 1R27 was produced. Such document which relates toa meeting held with the agents of the parties on 20.12.1988 showsthat the agent of the petitioner had agreed with the 2nd respondent’sview that the result of the election would not have been affected bythe failure to count the votes that would have been polled in thosepolling stations, in these circumstances I think that it is “notpermissible for the petitioner to make a complaint with respect to thefailure to conduct the poll at these 49 polling stations. In any event,even otherwise, the number involved being only 49 out of 8025polling stations in the island and there being no basis upon which tosay how the voters at these polling stations would have voted, thereis no material upon which to say that the result of the election forsuch electoral district would have been affected, or that at theelection the majority of the electors were or may have beenprevented from electing the candidate whom they preferred.
I would summarise then the effect of my findings thus 1
1.With respect to the petitioner’s case as contained in paragraph6A of her petition based upon general intimidation I hold that the
sc
Slrimavo Bandaranaike v. Ranasinghe Premadasa and Another
(S. B. Goonewardene, J.)
15S
petitioner has failed to establish the allegation that the majority of theelectors were or may have been prevented from electing thecandidate whom they preferred by reason of general intimidation asrequired by section 91(a) of the Presidential Elections Act.
With respect to the petitioner's case as contained in paragraph6B of her petition based upon the ground of non-compliance with theelections law I hold that the petitioner has failed to establish that theresult of the election was affected as required by section 91(b) of thePresidential Elections Act.
With respect to the petitioner's case as contained in paragraph6C of her petition based upon the claim of a failure to conduct a fairand free election in accordance with the provisions of the electionslaw, I hold that the petitioner has failed to prove that the majority ofthe electors were or may have been prevented from electing thecandidate whom they preferred as required by section 91(a) of thePresidential Elections Act.
With regard to the case urged for the petitioner at the stage ofaddresses that by reason of a large-scale breakdown of the electoralmachinery taken in conjunction with the acts of intimidationestablished, there was another circumstance shown to exist uponwhich the election was liable to be avoided, I hold that the petitioner
. cannot in any event succeed thereon on the basis that the majority ofelectors were or may have been prevented from electing thecandidate whom they preferred under the provisions of section 91(a)of the Presidential Elections Act.
With respect to the pleaded complaint that the poll at 49 pollingstations was declared null and void which was perhaps what wasreferred to by learned Counsel for the petitioner during his address, Ihold that the petitioner has not in any event established that the 2ndrespondent failed to conduct a fair and free election in accordancewith the elections law so as to say that the majority of electors were ormay have been prevented from electing the candidate whom theypreferred as required by section 91(a) of the Presidential ElectionsAct.
156
Sri Lanka Law Reports
[1992] 2 Sri L.R.
In the result I determine that the 1st respondent was duly electedand returned to the office of President and accordingly I make orderdismissing this petition.
There then remains the question of costs. The view I have beenconsistently holding, a view I did express during the hearing, wasthat the large and detailed volume of evidence led was only ofmarginal importance, and that view I believe is reflected in myapproach to the decision of this case. I therefore think that it mighthave been possible for the trial of this case to have beenconsiderably shortened and for the parties to have saved themselvesa good deal of expenditure. Taking that factor into account along withthe factor of the public interest element involved in the case, I wouldmake no order as to costs.
Election Petition dismissed with costs.Goonewardene, J. made no order as to costs.