The Standard of Proof For Crime in Civil Proceedings - A Ghanaian Perspective
The Standard of Proof For Crime in Civil Proceedings - A Ghanaian Perspective
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Cambridge University Press and are collaborating with JSTOR to digitize, preserve and extend
access to The International and Comparative Law Quarterly
A. Introduction
In English law it is well established that only two standards of proof are recog
nised. In criminal cases, the prosecution must prove the guilt of the accuse
beyond reasonable doubt. In civil cases the standard is the balance of probabilit
or preponderance of probabilities. Obviously the criminal standard is higher th
the civil one. Attempts to create a third standard lying between the criminal a
the civil standards have not been looked on favourably, particularly in England
Until 1957 it was not clear what the correct standard of proof was in civ
proceedings where a party alleged that a crime had been committed. On the on
hand, a line of authorities held that the appropriate standard was the crimi
one-beyond reasonable doubt." On the other, there were authorities which h
that the normal civil standard of balance of probabilities was more appropriate
In 1957 the English Court of Appeal had the opportunity to bring the debate
an end. It decided in favour of the civil standard-or, more accurately, a standa
1. Civil Suit No.328 of 1986, judgment delivered on 26 Jan. 1988 in the High Court o
Ghana at Koforidua. As yet unreported.
2. Idem, p.4 of the judgment.
3. (1982/83) G.L.R.D. 71, 74. In the unreported judgment the case is erroneous
referred to as In Re Agyepong (deceased) Palm v. Abosi.
4. Judgment on 30 Sept. 1991. Unrep. but see Independent, 1 Oct. 1991 and The Time
3 Oct. 1991.
5. See e.g. Dingwall v. J. Wharton (Shipping) Ltd [ 1961] 2 Lloyd's Rep. 213, 216 (per
Lord Tucker).
6. See e.g. The People of the State of New York v. The Heirs of Phillips [1939]3 All E.R.
952, a case alleging conspiracy to defraud; Willmett v. Harmer (1839) 8 C. & P. 695,
involving an allegation of bigamy in a libel suit; Issais v. Marine Insurance Co. Ltd (1923) 15
Lloyd's L.R. 186, where in an action on an insurance policy there were allegations of arson
on the part of the assured; and Thurtell v. Beaumont (1823) 1 Bing. 339.
7. See e.g. Hurst v. Evans 11917] 1 K.B. 352 (KBD) where in an action on an insurance
policy there was an allegation of theft on the part of the servant of the assured; and Doe d
Devine v. Wilson (1885) 10 Moo P.C.C. 502 (PC) where an allegation of forgery was made in
an action based on a deed.
Notwithstanding the existence of some cases where the point appears to have been
argued and decided in a contrary sense, I think the true view, and that most strongly
supported by authority, is that which the judge took, namely, that in a civil case the
balance of probability standard is correct.
And finally, in a statement which has been oft quoted, Morris LJ stated:"
But in truth no real mischief results from an acceptance of the fact that there is some
difference of approach in civil actions. Particularly is this so if the words which are
used to define that approach are the servants but not the masters of meaning. Though
no court and no jury would give less careful attention to issues lacking gravity than to
those marked by it, the very elements of gravity become a part of the whole range of
circumstances which have to be weighed in the scale when deciding as to the balance
of probabilities.
Two clear principles are discernible from the holding and these dicta. First, the
correct standard of proof in civil proceedings where a crime such as fraud is
alleged, is the applicable standard in civil actions generally-that is, proof on the
balance of probabilities-and not the higher criminal standard of proof beyond
reasonable doubt. Second, there is no absolute standard as such in either case.
Thus, the degree of probability must be proportionate to the subject matter. Or,
in the words of Denning LJ (as he then was), "the more serious the allegation the
higher the degree of probability that is required".
Hornal v. Neuberger, which settled the law in England, has been followed in a
number of cases. In In Re Dellow's Will Trusts,' where both spouses were found
dead in their home from coal gas poisoning, the issue before the court was
whether the husband or the wife had died first. As the evidence was not conclu-
It is interesting to note that Australia actually took the lead in this area of the
As far back as 1938, the High Court of Australia (the highest judicial bod
down the law in Briginshaw v. Briginshaw and Another." In that case,
appellant husband's petition for divorce on the ground of adultery was dism
by the trial judge because he was not satisfied that adultery had been p
beyond reasonable doubt. The High Court held that the appropriate stand
proof was the civil one. However, due regard was to be given to the seriousn
the allegation. As Dixon J (as he then was) pointed out:'9
It is often said that such an issue as fraud must be proved "clearly", "unequivoc
or "strictly" or "with certainty". This does not mean that some standard of pe
sion is fixed intermediate between the satisfaction beyond reasonable doubt req
upon a criminal inquest and the reasonable satisfaction which in a civil issue may
must, be based on a preponderance of probability. It means that the nature of
issue necessarily affects the process by which reasonable satisfaction is atta
When, in a civil proceeding, a question arises whether a crime has been commi
the standard of persuasion is, according to the better opinion, the same as upon
civil issues. But, consistently with this opinion, weight is given to the presumption
innocence and exactness of proof is expected.
The Australian position was put beyond argument by the High Cour
Australia in Rejfek and Another v. McElroy and Another.2" In that ca
appellants had sued the respondents for the rescission of a contract to purc
milk run in a suburb of Brisbane, and for damages, because of fraudulently u
representations inducing the contract. The trial judge, sitting in the Sup
Court of Queensland, gave judgment for the defendants because he wa
satisfied beyond reasonable doubt.
Having examined the leading authorities on the subject, the High Court
that, in the absence of statutory provisions to the contrary, proof in civil proc
ings of facts amounting to the commission of a crime has only to be establishe
a balance of probabilities. It stated: "No matter how grave the fact which is
found in a civil case, the mind has only to be reasonably satisfied and has no
respect to any matter in issue in such a proceeding to attain that degr
certainty which is indispensable to the support of a conviction upon a cr
charge.""2' The Court referred to its earlier decision of Helton v. Allen," in
17. The House had earlier approved Hornal in the divorce case of Blyth v. Blyth [1
All E.R. 524.
18. (1938) 60 C.L.R. 336.
19. Idem, pp.362-363.
20. (1965) 112 C.L.R. 517.
21. Idem, pp.521-522.
22. (1940) 63 C.L.R. 691.
23. (1989) 86 A.L.R. 257, 264. A decision of the Full Federal Court of Australia.
24. See Bond and Others v. Australian Broadcasting Tribunal (1989) 89 A.L.R. 185, 191.
This was confirmed by the High Court of Australia. For a discussion see N. L. Wallace-
Bruce, "How Are the Mighty Fallen-in the Most Obscure Places" (1991) 7 Queensland
University of Technology Law Journal 41.
25. Supra n.2.
26. Supra n.3.
27. In Re Agyepong (Decd.); Donkor v. Agyepong [1973] 1 G.L.R. 326, CA.
The learned judge found that P had been fraudulent in her application for grant of
letters of administration and therefore ordered the revocation of the grant. P
appealed to the Court of Appeal.21
The Court of Appeal allowed the appeal, concluding that the evidence estab-
lished conclusively that at the time of the marriage of P and A, the latter had
divorced all his other wives. Franqois JA, who dealt with the issue of the standard
of proof, stated the law in these words:-"
A grant obtained by fraud, malafide or by concealing information which should have
been disclosed to the court might also be revoked. However, fraud must not only be
pleaded but clearly and distinctly proved by the man who alleged it. Fraud in civil
proceedings required a higher standard of proof than an ordinary civil matter; there
must be proof of criminal deception. In the instant case, there could be no relief
unless the allegation of fraud to nullify the marriage under the Ordinance after the
death of one of the parties and also the allegation of fraud on the court in the
application for grant of letters of administration were proved strictly and clearly. The
circumstances surrounding the fraud must be inconsistent with any other view but
that of fraud.
His Lordship concluded: "In the result, the respondent and co-respondent failed
miserably to prove the fraud alleged and there could be no revocation of the grant
of letters of administration without such proof."'03 According to the Digest,
Franqois JA cited as his authority the dictum of Denning LJ (as he then was) in
Bater v. Bater.3
With all due respect to his Lordship, his statement of the law is contradicted by
the leading authorities both in England and Australia. In Bater v. Bater a wife
appealed a Commissioner's dismissal of her petition for divorce on the ground
that the charges of cruelty made by her were not proved beyond reasonable
doubt. A careful reading of the Court of Appeal's decision will reveal that Bater
cannot be relied on as authority for a general proposition that a crime such as
fraud requires "proof of criminal deception" in a civil proceeding.
Although the Court dismissed the appeal, the learned judges seemed to say that
the term "beyond reasonable doubt" was not used by the Commissioner with the
same strictness as it is used in the criminal jurisdiction. Denning LJ explained:32
If the Commissioner had, however, put the case higher and said that the case had to
be proved with the same strictness as a crime is proved in a criminal court, then he
would, I think, have misdirected himself because that would be the very error which
this court corrected in Davis v. Davis. It would be adopting too high a standard. The
divorce court is a civil court, not a criminal court, and it should not adopt the rules
and standards of a criminal court.
In a dictum which has since been frequently quoted in many cases,"3 but which
directly contradicts Frangois JA's formulation of the law, Denning LJ stated:34
It is curious that in In Re Agyepong (Decd.) the Ghana Court of Appeal did not
advert to the cases of Hornal v. Neuberger Products Ltd and Reffek and Another
v. McElroy and Another. As we have already seen, those cases not only clarified
but indeed settled the law in England and Australia respectively. Moreover, they
have been followed subsequently in a number of cases.
Whilst cases decided in England, Australia and elsewhere are not automatically
binding on the Ghanaian judiciary, they have strong persuasive force. This is
especially so with English precedent, which invariably becomes the law in Ghana
unless there is contrary local legislation on the particular subject. Apparently, this
is the case throughout the British Commonwealth.
Reverting to our discussion, it would appear that the law as stated by the Ghana
Court of Appeal is a misreading of English case law rather than an interpretation
of local legislation. Serious doubt must therefore be cast on the value of In Re
Agyepong (Decd.) as authority in Ghanaian law on the issue of the correct
standard of proof in civil proceedings where there is an allegation of criminal
conduct such as fraud, arson or even murder.
E. Conclusion
Third, although the judge's formal adoption of the criminal standard is open
criticism, and certainly contrary to the leading authorities on the issue, his act
application of the standard is perhaps not far off the mark. He stated:31
I have proceeded, as indeed Mr Scrivener invited me to, on the basis that no on
whether in a criminal or a civil court, should be declared guilty of murder, certain
not such a terrible murder as this, unless the tribunal were sure that the evidence did
not admit of any other sensible conclusion.
In other words, if one puts aside for a moment the correct terminology, Rougi
J's approach was to be "sure" of the evidence. This is not inconsistent with wh
has been said in the cases discussed in this article, but with a proviso that t
standard should not reach the criminal standard. For example, as Denning
stated in Hornal: "The more serious the allegation the higher the degree
probability that is required: but it need not, in a civil case, reach the very hig
standard required by the criminal law."''3 Similarly, in the Australian case
Briginshaw v. Briginshaw and Another Dixon J reminded us that, even if the cour
uses such terms as "certainty", "strictly" and the like, the standard should still
rise to the criminal standard.4"
In the case itself, the judge had to decide between only two stories. His task w
not so much to determine whether there had been a murder, but rather to cho
37. Ibid.
38. Ibid.
39. Supra n.9.
40. Supra n.19.
* Barrister and Solicitor (Australian Capital Territory, Ghana and Victoria); Solicitor
(New South Wales and High Court of Australia); Lecturer, Department of Management,
University of Western Australia.