The republic of uganda in the supreme court of uganda at kampala

Burden and standard of proof:

Both the Constitution and the Presidential Election Act, to which I shall hereinafter refer no “the Act”, provide that this Court shall inquire into and determine expeditiously, the petition challenging the result of a Presidential election, without prescribing the mode of inquiry. Pursuant to S.58(1 1) of the said Act, rules of procedure for the conduct of such a petition were made, prescribing an adversarial mode of inquiry for the petition, similar to the trial of civil cases, except that evidence has to be affidavits. Counsel therefore, addressed the Court on the questions of burden and standard. It was basically common ground that the burden of proof lay on the Petitioner. However there was no consensus on whether there was any shift of evidential burden in this case. I will discuss that under issue No.4 where it is more particularly relevant. There was also no consensus on the standard of proof required in election petitions. It was only common ground that it is higher than a balance of probabilities. I will express my views on this subject at this juncture, because it affects all the principal framed issues. The argument on standard of proof in the instant case arises from the provision in S.58 (6) of the Act that a ground for annulment of the election of a candidate as President, has to be “proved to the satisfaction of the Court.”

In their respective submissions, counsel revealed considerable divergent views on the matter particularly in regard to comparison with the standard of proof in criminal cases. For the Petitioner, Mr. Balikuddembe argued that the expression “proved to the satisfaction of the Court”, set the standard of proof well below that required in criminal cases. He contended that it was slightly above that of a balance of probabilities, and equated it with that required to prove fraud in civil cases. For the 1st Respondent, two views were expressed. Mr. Bitangaro, submitting on the law applicable to issue No.4, submitted that the standard to apply was proof beyond reasonable doubt, citing as authority BARTER vs. BARTER (1950) 2 All ER 458; MBOWE vs. ELIUFOO (1967) EA 240, and MARGARET ZIIWA vs. NAVA NABAGESERA Civil Appeal No.39/97 (C.A) (unreported). Dr. Khaminwa, however, modified this when submitting on issues Nos. 2 and 3 by contending that the standard was below that required in criminal cases but was very high. He also cited BARTER vs. BARTER (supra) and MBOWE vs. ELIUFOO (supra) and added KATWIREMU vs. MUSHEMEZA Election Petition No.1/96 (H.C. Mbarara) (unreported). For the 2nd Respondent, the learned Solicitor General put it dramatically saying that the standard of proof in election petitions was so high that “its leaves touched the under-belly of the standard in criminal cases.

It appears to me that there are two possible approaches to the question. One is to consider the statutory expression “proved to the satisfaction of the Court” as fixing the level of proof required. The other is to consider that by that expression the legislature left it to the Court to determine what amount of proof would satisfy it that the matter in issue has been proved. The Courts both in Uganda and outside have considered the same or similar expression. I need only refer to a few.

Let me start with decisions of the English Courts which have had persuasive influence on the interpretation put on the expression by the Ugandan Courts. They considered the statutory expression “if the Court is satisfied” in regard to proof of matrimonial offences. In BARTER vs. BARTER (supra) all three Lords Justices of the Court of Appeal agreed that where the Court had to be satisfied in regard to cruelty in a divorce case, it was not a misdirection to state that the Petitioner must prove her case beyond reasonable doubt. Bucknell L.J. said:

“I do not understand how a Court can be satisfied that a charge has been proved (and the statute requires that the Court shall be satisfied before pronouncing a decree) if, at the end of the case it has a reasonable doubt in its mind whether the case has been proved. To be satisfied and at the same time to have a reasonable doubt seems to me to be an impossible state of mind.”

The House of Lords in PRESTON-JONES vs. PRESTON JONES (1951) AC 391, considered the same expression in relation to proof of adultery in a divorce case. It endorsed the decision in BARTER vs. BARTER (supra). Lord MacDerrmott said:

“I am unable to subscribe to the view which, though not propounded here, has had its adherents, namely, that on its true construction the word ‘satisfied’ is capable of connoting something less than proof beyond reasonable doubt. The jurisdiction in divorce involves the status of parties, and the public interest requires that the marriage bond shall not be set aside lightly or without strict inquiry. The terms of the statute recognise this plainly, and I think it would be quite out of keeping with the anxious nature of its provisions to hold that the Court might be ‘satisfied’ in respect of a ground for dissolution, with something less than proof beyond reasonable doubt.”

However, in BLYTH vs. BLYTH (1966) AC 644, The House of Lords, while not departing from its earlier decision in the PRESTON-JONES case (supra), focused on contrast between proof of a matrimonial offence as a ground of dissolution of marriage and proof of absence of condonation. By majority it was held that although in either case the Court must be satisfied, a lesser degree of proof is required to satisfy the Court that the presumption of condonation has been rebutted, than would be required to satisfy it that a matrimonial offence has been committed. The majority view was that by the use of the phrase “Court if satisfied” the legislature did not thereby fix a standard of proof but left it to the Court to determine the quantum or degree of proof that will “satisfy” it, depending on the gravity of the subject matter.
The subject matter in the instant case is the validity of election of the President. Its gravity cannot be disputed. The statutory expression under consideration is “to the satisfaction of the Court.” MBOWE vs. ELIUFOO (supra) was an election petition in the High Court of Tanzania, in which the expression “proved to the satisfaction of the Court” in the applicable Tanzania statute was considered. In the judgment of the Court, Georges C.J., agreed with the approach in BARTER vs. BARTER (supra) and held that:

“the standard of proof in this case must be such that one has no reasonable doubt that one or more of the grounds set out in S. 99(2) (a) has been established.”

The High Court of Uganda in many decisions handed down since the first post- independence parliamentary elections in 1980, has followed MBOWE vs. ELIUFOO (supra). They include, the following election Petitions: CLEMENT TIBAROKORA vs. R.O. RUKUNGIRI &ANOTHER No. MKA 1/81; Z.C.ILUKOR vs. R.O. &ANOTHER, No.MM 1/89; ODETA vs. OMEDA No.NP 1/96; AYENA ODONGO vs. BEN WACHA No. 2/96 (Gulu) (all unreported). In its judgment in R.O. KAMPALA, MARGARET ZIIWA & ANOTHER vs. NAVA NABAGESERA (supra), the Court of Appeal referred to those decisions with approval and singled out for review the decision of the Principle Judge in ODETA’s case with which the Justices of Appeal it agreed and then concluded:

“The effect of the holding in the Mbowe case and the Uganda cases that have followed that decision is that grounds for setting aside an election of a successful parliamentary candidate set out in S.91 of Statute No.4 of 1996 must be proved beyond reasonable doubt. This is because the Court cannot be satisfied if there was a reasonable doubt.”

In the KATWIREMU Case (supra) decided before this decision of the Court of Appeal, Musoke-Kibuuka J., expressed a view which appears not to have been adverted to by the Court of Appeal, in its decision I have just referred to. The learned Judge expressed the view that if Parliament had intended proof of election petitions to be beyond reasonable doubt, it would have said so expressly, and would not have provided for separate criminal trial of allegations proved as electoral offences in the election petition. He took a stand as follows:

“....everyone seems to be agreed that, whatever name is given, the standard of proof required for an allegation to be proved to the satisfaction of the Court under S.91(1) of the Parliamentary elections (Interim Provisions) statute 1996, is proof which is higher than that which is required in ordinary civil suits. That, in my view is sufficient for disposal of the allegations made in this petition.”

In S.58 (7) and (9) of the Act it is expressly provided that when hearing an election petition, this Court has no power to convict a person for a criminal offence, and that where it appears from the facts that a criminal offence may have been committed the Court shall make a report to the DPP stating the nature of the offence and the name of the person (who committed it) for appropriate action. To my mind these provisions are not an indicator that Parliament intended the standard of proof to be lower than beyond reasonable doubt.- If at all the provisions be a reflection on Parliament’s intention on the standard of proof, then, in my view, the more plausible interpretation would be that, Parliament realising that it had set up a high standard of proof, (equivalent to that in criminal cases), made it clear in those provisions that it was not thereby conferring on the Court power to convict any person of a criminal offences. The reason is obvious. It is recognition of the fact that our jurisdiction does not yet allow for a criminal trial within a civil trial. It is only in the recent past that our legal system introduced the possibility of an award of a civil remedy in a criminal trial. The reverse has not yet been introduced.

Furthermore, in my view it does not follow that the fact that Parliament did not use the expression “beyond reasonable doubt”, it could not have intended the equivalent. The expression it used is not inconsistent. The learned judge appears to hold the view that in using the expression “to the satisfaction of the Court” Parliament set up yet another standard of proof at a level between that of “balance of probabilities “and that of “beyond reasonable doubt.” I respectfully do not share that view. I think the reason with which I agree, why the Courts have sought to fit the expression within the known standards is a desire to have an objective test as to what amount of proof ought to satisfy the court in such cases, rather than leaving it to be subjectively determined in every case. I do share the view that the expression “proved to the satisfaction of the Court” connotes absence of reasonable doubt. Admittedly the word “satisfied” is adaptable to the two different standards. It is not uncommon for a court to hold that it is “satisfied on a balance of probabilities”, or that “it is satisfied beyond reasonable doubt.” However, where the Court holds that it is satisfied per Se, that a matter has been proved, or that a matter has been proved to its satisfaction, without more, then to my mind there can be no room to suppose that the court harbours any reason doubt about the occurrence or existence of that matter. By requiring that the ground for annulment of an election be proved to the satisfaction of the Court, the legislature laid down the minimum amount or standard of proof required. The amount of proof that produces the Court’s satisfaction must be that which leaves the Court without reasonable doubt.

To my mind it does not matter that proof in criminal cases is also required to be beyond reasonable doubt. The law provided a high standard of proof for criminal offences because of the grave consequences of a criminal conviction. Similarly, because of the gravity of annulment of an election, the law provided for a high standard for proof of the grounds of annulment. I found support for this view in what Lord MacDermott said in PRESTON-JONES Case (supra). After stressing that the standard of proof for dissolution of marriage could not be less than proof beyond reasonable doubt, he said:

“I shall perhaps add that I do not base my conclusion as to the appropriate standard of proof on any analogy drawn from the criminal law. I do not think it is possible to that the two jurisdictions are other than distinct. The true reason, as it seems to me, why both accept the same general standard — proof beyond reasonable doubt-lies not in any analogy, but in the gravity and public importance of the issues with which each is concerned.”

In BLYTH vs. BLYTH (supra) Lord Morris of Borth —y- Gest said:

“…..the jurisdiction in divorce is statutory and by statute certain duties are imposed upon the Court. There is no occasion to seek to compare or to equate the jurisdiction in divorce with jurisdiction in either criminal or in other civil matters.”

In a similar vein I would say that it is unnecessary to seek to compare or equate the jurisdiction in election petition, with jurisdiction in either criminal or in other civil matters. In conclusion from both the import of the words used in the statute and the gravity of the subject matter, I would uphold the long standing view which the Uganda Courts have held that the standard of proof must be that which leaves no reasonable doubt in the Court’s mind.