G. Vagheesam Pillai vs M.R. Krishnamurthi Padayachi and Another

Anantanarayanan, J.—On 19th February 1962, elections were held for the Kattumannarkoil Assembly Constituency, in Madras State. After counting of votes on 25th February 1962, M.R. Krishnamurthi Padayachi (first Respondent), who contested the election as a candidate of the Dravida Munnetra Kazhagam was declared elected, with a count of 27,706 votes in his favour. Vagheesam Piilai, the Appellant before us, who contested as a Congress candidate, polled 27.599 votes, with a difference of 107 votes in favour of the first Respondent. The Appellant then instituted an election petition challenging the validity of this election, under Sections 100 and 101 of the Representation of the People Act (XLIII of 1951). It is not in dispute before us that there was no ground of objection that the election was vitiated by any corrupt practice committed by the returned candidate or his election agent, or by any other person with the consent of either of them, Section 100(1)(b). On the contrary, it has throughout been conceded that the Appellant was never able to obtain material for such an averment, or even for reasonable suspicion in that respect. The grounds actually advanced related to Section 100(3)(ii) and (iii); it was alleged that the result was materially affected by the admission of 272 votes, which were instances of impersonation of voters, either dead on the date of election, or who had been detained abroad, or who were absent from the station (Lalpet) on the day of poll, though in India. After a recount and scrutiny ordered by the Tribunal, the difference between the parties was reduced to 90 votes.

There was then a very elaborate enquiry, and the order in the election petition runs to 157 printed paragraphs. The Appellant examined P.Ws. 1 to 98 and the first Respondent R.Ws. 1 to 33; a large number of documents (exhibits A-1 to A-148) were marked for the Appellant and 21 documents (exhibits B-l to B-21) for the first Respondent. Ultimately, after an elaborate discussion of the evidence, with regard to each individual instance, seriatim, the Tribunal came to the conclusion that there was impersonation of dead or absent voters, in certain cases. Thirty-five votes were then declared void, in the light of the findings on issues 3 to 5. Deducting those votes, the difference between the two candidates was reduced to 55. But indisputably, that did not materially affect the result of the election, and the petition was dismissed.

The matter has been elaborately argued before us by learned Counsel on both sides. At the outside itself, we desire to take note of certain arguments by Sri Venugopal for the Appellant, regarding the true perspective of approach to the facts and the evidence, in the light of certain legal principles. The Tribunal did place some reliance on the fact that the allegations virtually amounted, not merely to corrupt practices by the concerned impersonates, whoever they were, but even to acts, which would be punishable as offences under the Indian Penal Code. The Tribunal then observed, following M.A. Muthiah Chettiar v. Saw. Ganesan 21 E.L.R. 215 that such an enquiry is really of the character of a quasi-criminal trial, that an election Petitioner, under such circumstances, was virtually a prosecutor, and that the case must be made out by positive proof, not by conjecture. A Bench of the Andhra High Court also observed in V.B. Raju v. V. Ramachandra Rao 21 E.L.R. 1 that charges of corrupt practice were quasi-criminal in character, and that the allegations thereto must not merely be clear and precise, but have to be established by evidence of a conclusive nature, in view of the fact that penal consequences flow from a disqualification arising from a corrupt practice. The argument of learned Counsel for the Appellant is that this approach of the Tribunal is fallacious, because the present petition does not really involve any averment of a major corrupt practice under Sub-section (3) of Section 123 of the Act; had that been the case, even a single instance of corrupt practice would vitiate the election. Nevertheless, as learned Counsel for the first Respondent has rightly urged, it appears to us that the fact remains that the averments are those of deliberate impersonation of dead or absent voters, by others, obviously, under the influence of interested adherents. While it is true that nothing is urged either against the returned candidate, or his election agent, we must necessarily reckon with the fact that the averments, if proved, are grave, constituting offences under the criminal law.

In this setting of facts, Sri Venugopal has argued that the Tribunal misdirected itself in two vital respects. Firstly, the standard of proof cannot be the same as that which has to prevail in a criminal prosecution; on the contrary, there are definite authorities for the view that a preponderance of probabilities would suffice, and not the establishing of guilt beyond reasonable doubt. The authorities, particularly in England, are definite that even where an allegation of criminal conduct has to be proved, as affecting some third party, in a civil action, preponderance of probabilities will be enough. Further it is urged, on the authority of Sudhansu Sekhar Ghosh v. Satyendra Nath 4 E.L.R 73 (West Bengal Tribunal) that not merely does the onus, shift on, prima facie proof by the Petitioner, but that, in election petitions at least, there is an onus on the Respondent also; another relevant authority to the same effect is Abdul Rouf v. Makhtar Ali 2 E.L.R. 340 (Gauhati Tribunal).

We think it is important that the arguments on this aspect should receive some initial scrutiny at our hands before we proceed to the facts. In Halsbury''s Laws of England (Simonds edition), volume XIV, paragraph 384, with regard to proof of a corrupt practice, like bribery, the dictum is set forth that proof may be circumstantial, and that the Court is not bound by the strict practice applicable to criminal case. The dual standard, as between criminal and civil actions, has received the fullest treatment in Hornal v. Neuberger Products, Ltd. (1956) 3 All E.R. 970 which may be stated as the leading authority on the subject, Denning L.J., observed:

The more serious the allegation the higher the degree of probability that is required; but it need not, in a civil case, reach the very high standard required by the criminal law.

The learned Judge then referred to Bater v. Bater (1950) 2 All E.R. 458 and Lek v. Matthews (1927) 29 Lioyds. Rep. 141. Hodson L.J. reviewed the case law from Cooper v. Slade 6 H.L.C. 746 onward, and referred to the dictum of Professor Kenny that, in a civil action, a preponderance of evidence will be sufficient, and that proof beyond reasonable doubt is really not required. But as Denning L.J., observed in Bater v. Bater (1950) 2 All E.R. 458 there may be degrees of probability within the standard of proof of a civil action, the degree depending on the subject-matter. The entire subject has been summarised in the standard treatise of Cross on Evidence (1958 edition, page 124) and the result of the discussion is expressed in the following form:

Although there were several previous decisions which were not discussed by the Court of Appeal, Hornal�s Case (1956) 3 All E.R. 970 may be taken to have settled the English Law for the time being. An allegation of criminal conduct need only be established on a preponderance of probability, in a civil action. It is submitted that in every case the question is whether there is any compelling reason for adding the gloss that, though the Tribunal of fact may consider an issue to be proved, they must find the contrary if they entertain a reasonable doubt.

This reference to the state of law in the United Kingdom will not be adequate, without some reference to the different principles that might apply in this country, resulting from the statutory enactment of the law of evidence. As is well-known, the Indian Evidence Act does contain a definition of when a fact may be said to be proved; in Section 3 of the Act, the standard is that of a prudent man who considers, on the evidence, the existence of the fact so probable, that he is led to act on that supposition. In Sarkar on Evidence (tenth edition), the cases upon the different standard in civil and criminal causes respectively will be found summarised at page 30. Bater v. Bater (1950) 2 All E.R. 458 is referred to, as well as Hornal�s Case (1956) 3 All E.R. 970 Narasingamuthu In re ILR (1949) Mad. 357 is authority for the view that in a quasi-criminal proceedings, prima facie proof may be adequate to shift the onus to the other side.

This discussion of the case law leads us to the following propositions. In this country also there is foundation for the broad principle that where what might constitute a criminal offence has to be proved as relevant, in a civil proceeding, it may well be upon a preponderance of probability; proof beyond reasonable doubt need not be insisted on. Further, there is certainly a shifting of the onus to the opposite party, where the facts are strong enough to justify the inference of a clear prima facie case. Again, it is true that, particularly in an election case, where the purity of the electoral process is the prime consideration, the Respondent, namely, the successful candidate is not without his obligations. He is bound to adduce such evidence as lies in his power; but this would be on stronger ground where a corrupt practice, vitiating the election, is alleged against him, or his election agent, or against some one with the connivance of either. Per contra, we are equally clear that where a grave averment, like the impersonation of a dead or absent voter, is alleged, mere suspicion or conjecture, or hazy evidence, will not do. The facts remain that there is a ballot paper, apparently relating to the voter, that the election officer issued that paper to the claimant, and that, ordinarily, candidate will have his election agent, or authorised representative, in some booths at least, if not in every booth, to make a challenge in a dubious case. On this aspect the facts are heavily against the Appellant. There is a long story in the record of how the authorization issued to the election agent by the Appellant, for Lalpet, was snatched away by certain opponents, which led to an actual police complaint, but we are quite unable to judge the veracity of this explanation. It is certainly extraordinary that, with respect to no single booth in Lalpet, was there some one authorised on behalf of the Appellant, to challenge the very large number of alleged impersonated votes. While commenting on individual cases, we shall notice the argument that, in some instances at least, according to the learned Counsel for the Appellant, the onus shifted to the first Respondent at a particular stage. But if regard is paid to the fact that the ballot papers do exist in the names of the voters, and that there was hardly any challenge of any of these votes at the time of the actual polling it will be clear that, even for the onus to shift to the Respondent, there must be fairly definite evidence of impersonation. It is true that the Tribunal found that there were 35 votes void for impersonation; but that will justifiy no argument that such corrupt practices might have occurred to a greater extent.

There is another aspect of the matter, which is equally important, and which might be immediately referred to. Sri Venugopal argues that, where the entire evidence is before Court, and it is conflicting in character, the matter of onus is really academic; vide Narayan Bhagwant Rao Gosavi Balajiwale v. Gopal Vinayak Gosavi (1960) 1 S.C.R. 773; Moran Mar Basselios Catholicos v. Avira ILR (1958) Ker. 134 (S.C.) Parasnath Thakur v. Mohani Dasi and Ors. (1960) 1 S.C.R. 271. In Nirmal Kumar Nawlakha v. Sant Lal Mahto (1937) ILR Pat. 622 the dicta of Lord Hanworth M.R. in Stoney v. Eastbourne Rural Council (1927) 1 Ch. 367 have been cited with approval; the principle is that the party on whom the onus lies, in the first instance, must establish a sufficient case to justify a judgment in his favour, if there is no evidence per contra.

The other aspect that we might immediately refer to is that, by and large, this election case depends on the appreciation of conflicting oral testimony. As we shall later show the documentary evidence is really of very little assistance in any of the categories in which the alleged instances of impersonation fall. Undoubtedly, in the appreciation of such oral evidence, the Tribunal had the very great advantage of seeing and hearing the witnesses ; even if the discussion in the judgment does not relate to the demeanour of witnesses in every case, or most of the cases, that advantage is not to be treated with slight regard. Actually, in dealing with an appeal u/s 116-A of the Representation of the People Act, such findings of fact recorded by the Tribunal, on an appreciation of oral evidence, are entitled to great respect. A mere possible contrary view, would not justify us in coming to an opposite conclusion, in any particular case; further it is not the law that merely because some witnesses have come forward to speak to a particular fact, more especially a negative fact, such as the absence of a voter at an election, they should be believed: vide the dicta in Jamuna Prasad Singh Vs. Shri Ramnivas and Others, .

It is true that it is certainly open to this Court, where the facts justify it to arrive at a different conclusion, even in a matter that depends purely on the appreciation of oral evidence, and even where we have not had the benefit of seeing and hearing the witnesses. As observed by Viscount Simon in Wait or Thomas v. Thomas (1947) A.C. 484, 486 a judgment of first instance is not infallible, but the view of the trial judge as to where the credibility lies, is entitled to great weight. In Renmax v. Austin Motor Co. Ltd. (1955) 1 All. E.R. 326 Viscount Simonds made a distinction between the question of the credibility of witnesses, and the question of the inference to be drawn from specific facts. In the latter instance, the Appellate Court is in as good a position to evaluate the evidence, as the trial Judge. But on a question of fact depending upon oral testimony, as is the case in almost every one of the instances before us the preponderence of authority is weighily to the effect that an appellate Court should not lightly differ. We think it is sufficient to refer to the authorities very briefly. They are Sarju Pershad Vs. Raja Jwaleshwari Pratap Narain Singh and Others, ; Radha Prasad Singh v. Gajadhar Singh and Ors. (1960) 1 S.C.R. 663; AIR 1948 20 (Privy Council) ; Veeraswami v. Narayya I.L.R (1949) Mad. 487 (P.C.) and Jamuna Prasad Singh Vs. Shri Ramnivas and Others, .

Having so far outlined the principles that must underline our perspective of approach, we might proceed to discuss the categories into which the evidence falls. Of the votes arraigned as those of impersonation of dead persons, eleven have been held against the Appellant and we shall review those cases that appear to be strongest for the Appellant. The second category consists of persons claimed to be absentees from the polling station that day but in India, and the third of students of the local Arabic College similarly claimed to be absent, being in their villages. Fourteen instances of the second group have been held against the Appellant. We shall note certain of those instances. Of the Arabic College students, 28 have been held against the Appellant, with regard to alleged acts of impersonation. Of those claimed by the Appellant to have been absent abroad, in such places as Singapore and Malaya, 56 have been held against the Appellant. We might immediately here clarify the matter of double voting. The Tribunal found that, in five cases, votes were polled in two different polling booths; they were rejected accordingly. The argument is that even in the remaining cases, double voting is a fact, as shown by the records, and that the votes are void by reason of Section 62(3) of the Representation of the People Act. Accepting this argument, it is obvious that the margin would be reduced by a further five votes but no more. We shall now proceed to scrutinise certain of the instances under the categories outlined above, with remarks generally applicable to those categories.

With regard to votes of persons claimed to be dead, it is clear that the extracts from the register of deaths would have been the best evidence. In each one of those Cases, the ballot paper is there as prima facie evidence of the voter having exercised his franchise. Here, we must reiterate that it is surprising that not a single authorised agent on behalf of the Appellant seems to have challenged any of these votes at any of the booths. We are by no means satisfied regarding the story about the police report; and the alleged frustration of the election agent of the Appellant, by some incident of rowdyism on a previous day, is hardly any adequate explanation. On the individual instances, with regard to Mahamed Ismail (paragraph 23), it is pointed out that P.W. 26 is corroborated by the village Munsif (P.W. 11), which evidence the Tribunal ignores. But the death extract is not produced, and in exhibit A-14, the name of the father is not given. With regard to Kolandaivelu (paragraph 32) it is pointed out that a Hindu woman, like P.W. 25 is hardly likely to claim that her husband was dead, unless it was a fact. But the Tribunal seems to be right in the statement that the extract from the hospital register at Mannargudi could have placed the matter beyond doubt. The husband of P.W. 25 might have been a different person from the voter, and the corroboration by P.W. 55 seems to be of little value. With regard to P.W. 39 (paragraph 33) shaddagar of the alleged dead voter, his evidence was not believed. With regard to P.W. 10 (paragraph 37) in this case alone we might consider the onus as having shifted to the first Respondent, in view of exhibit A-38 (death extract), though the nexus between this and the voter is not very clearly established. With reference to instances in paragraphs 34, 35, 36, 39, 41, 42 and 43 of the judgment of the Tribunal, the last alone would appear to deserve some mention, as the husband (R.W. 17) admitted the death of the voter though, according to him, this was in April-May 1962. Thus even if we take two of the instances in this category as established for the Appellant, this would reduce the margin only to a very limited extent.

We may now refer to the next category, absentees in India, other than students. We have been taken through this evidence in elaborate detail. In one of those instances alone that of the special deputy tahsildar, whose case is dealt with in paragraph 56, we are inclined to hold that the evidence has to be accepted, as he is a responsible officer against whom no motive is alleged for giving a false deposition. In all the other cases, the evidence appears to us to be quite inconclusive, and insufficient to shift the onus. The ballot papers are there, and the impossibility, of exercise of the franchise, through absence elsewhere, has by no means been established in any instance. The fact that the voter has given evidence, in several of these instances, is, in our view, both misleading and unsafe as a criterion for any definite inference. It is at least as conceivable that a person having exercised his franchise in one way, may be later won over to depose differently, as that some other person, in the same village, might have been persuaded to impersonate him, risking prosecution for a criminal offence. The absence of evidence in rebuttal in several cases has been stressed by Sri Venugopal. But, where the oral evidence did not appear to be worthy of credit, it is difficult to perceive how their feature can lead us to the accepting of that evidence. On both sides, there are one or two omnibus witnesses. Obviously, such evidence has to be accepted with great caution, particularly where that kind of witness for the Appellant gives evidence that, on the date of election, a number of voters catalogued together'' were not in the village, and did not exercise their franchise.

With regard to the next category, namely, students of the Arabic College a local religious institution, the probabilities are not at all in favour of the Appellant. The attendance register, exhibit A-1, produced through the principal of the college, is a most unsafe basis for any inference. The Tribunal treated exhibit A-1 as inadmissible, relying upon Section 34 of the Indian Evidence Act. The argument is that provision of law can apply only to account-books, wherein self-serving statements are to be found, and sought to be used for foisting monetary liability on third person. But, assuming that the attendance register is admissible, there is no evidence of the teacher or teachers making the entries. From 22nd January to 23rd March 1962 was a period of recess for the College and the poll was on 19th February 1962, in the middle of this recess. It is true that students were likely to avail themselves of the recess in order to visit their parents in neighbouring towns and villages. It is further true that students marked as absent for a few days before the commencement of the recess, or after the reopening, are definitely, unlikely to have continued to stay in the hostel throughout. Per contra the record clearly shows that the students could have stayed in the hostel, during the recess. Further, in most of these cases, the students could have proceeded if at all, to their native villages, within a radius of 100 to 200 miles. Considering the nature of a political election in India today and exertions made to secure votes, as also the excitement in the electorate with regard to the privilege of franchise, it is not at all improbable that many of these students could have returned just prior to the election day, to Lalpet, to exercise their franchise. In several cases, the concerned students have given evidence, but this does not shift the onus at all, since the ballot papers are prima facie evidence of the exercise of the franchise. We find it difficult to imagine that, in so many instances, there were other young persons readily available to make false personation and that, in no conceivable case, was this challenged by any one on behalf of the Congress candidate.

Again, with regard to the next category, those said to be absent abroad particularly at Singapore and Malaya, the travel registers relied on by learned Counsel for the Appellant, exhibits A-49 to A-85, and the embarkation and disembarkation cards, exhibits A-86 to A-1, furnish, in our considered view, no safe basis for the inference of absence abroad, and an impersonation by another. We have to take judicial notice of certain inferences, suggested by the facts of the record. Many persons in this area, which seems to be predominately that of a Muslim population, make frequent journeys to Singapore and Malaya on business. They stay for business. They stay for indefinite periods, and return with equal frequency. The travel registers do show departure to places abroad, but the Tribunal was right in confining the improbability of a return of the voter, to instances where the departure had taken place very recently prior to the election. The travel registers are quite inconclusive, for they relate only to journeys by two named vessels from Madras Harbour. There is evidence that Nagapattinam is an equally possible place of disembarkation, and there is evidence that travel by air also is feasible. For that matter, even Visakhapatnam Port could have served and there is absolutely no warrant for a presumption that when a resident of these localities proceeds to Malaya or Singapore, he is likely to continue there indefinitely, and not to return; the inference should be the other way round. We must reiterate that, in none of these cases, is the absence of the voter established by any thing except oral evidence. There can be no presumption u/s 114 of the Indian Evidence Act that because a person left for Malaya or Singapore a considerable time previously, he could not have returned to participate in the election at Lalpet. The absence of rebuttal evidence can be a basis for criticism, only where the evidence is definite, clear and strong enough to justify a prima facie inference of the absence of the voter abroad, and of the exercise of the franchise being due to impersonation. Otherwise, the onus does not shift. In all these cases, we are not satisfied that we would at all be justified in differing from the detailed conclusion of fact recorded by the Tribunal, after a very elaborate appreciation of the oral evidence.

In brief, therefore, we are convinced that the margin cannot be reduced so as to materially affect the result of the election, by any conceivable evaluation of the evidence, even on a standard most favourable to the Appellant. Even upon such a standard, and taking the individual instances most favourable to the Appellant, in his favour, in the several categories, we have found ourselves unable to reduce the margin to below 40 or so. It is not in dispute that, even if the margin is a single vote, in favour of the first Respondent, the appeal would have to be dismissed and the ultimate finding of the Tribunal upheld.

We have been at some pains to examine the aspect whether, in view of certain corrupt practices, which are proved to ha?& occurred at this election, though the candidate or his agent could not at all have been responsible, we should interfere, because of our concern to safeguard the purity of the electoral process. Reference may be made, in this context, to the dicta of the Supreme Court in Jagan Nath Vs. Jaswant Singh and Others, ; also see the dicta of a Bench of the Punjab High Court in Sadhu Singh Vs. Shamsher Singh Josh and Others, . It is true that Courts have a concern in this matter, beyond the mere objective of a just decision upon the facts, as between the parties. Election is essentially a part of the democratic process, and the Courts are also custodians of the purity of that process, to the extent to which the Legislature has entrusted them with safeguarding powers. But, obviously such considerations will have no relevance, where it is not a corrupt practice per se which vitiates the election, but the invalidity of certain votes; it must then be established that the result has been materially affected; otherwise, the Courts are powerless to interfere. But the state of affairs disclosed by the records does show that abuses can easily occur. It would appear that the electoral rolls, though prepared well in advance are not prepared with adequate safeguards against errors, such as the inclusion of the names of dead persons, or persons permanently residing abroad; the mistakes might well have been bona fide; but such features indicate that authorised agents of political parties, and candidates themselves, should not content themselves merely with participating in elections. The accurate preparation of the electoral rolls, should receive the attention of all responsible political parties and their agents, if the democratic process is to be pure.

In the circumstances, the appeal is dismissed; the parties to bear their own costs here.