Kelly, R. v

In 1960, Parliament by enacting the Betting and Gaming Act 1960 made significant changes to many centuries of legislation in respect of betting and gaming. The Act repealed many old Acts of Parliament, gave a new definition to “gaming” and established a licensing regime for those who provided premises for gaming. Those and other provisions were consolidated into the Gaming Act 1968. S.52(1) of that Act (substantially re-enacting s. 28 of the 1960 Act) defined “gaming”, subject to provisions that are immaterial, as

Ss.3 and 4 of the Act prohibited levying a charge in respect of gaming or a charge on stakes or winnings, unless the premises were licensed; s.8 provided that if gaming took place contrary to the prohibitions, an offence was committed. The issue on this appeal is whether the judge correctly directed the jury in respect of the statutory definitions of a game of chance in circumstances where the appellant organised a specific type of poker game at unlicensed premises.

There has been no decision which can be found on the meaning of the provision defining a game of chance since the change in the law over 40 years ago. It is not clear why this is the position. It has proved very difficult to ascertain whether there have been many prosecutions, as no records were kept where there was an acquittal and convictions were deleted after a given period of time. We are grateful to counsel for the Crown for ascertaining in these difficult circumstances that since the change in the law in 1960, it can now only be established that the first recorded conviction was in 1998 and thereafter there have been a few convictions each year. The Report of a Joint Committee of the Lords and Commons published in April 2004 (HL Paper 63-I, HC 139) recorded that the Gaming Board and the police acknowledged difficulties in tackling illegal gaming due to a lack of police expertise and police time.

In 2005 Parliament amended the provisions again by the Gambling Act 2005; a game of chance in s.6 of that Act is defined:

The circumstances in which the issue now arises for decision can be briefly described.

In March 2004, the appellant opened the Gutshot Club in Clerkenwell Road, Islington, London where he organised a variant of poker called “Texas Hold ‘Em Poker” (TH Poker); he charged players for participation and made a levy on the winnings. The participants in TH Poker were largely enthusiasts for the game and the evidence before the courts was that all the profits in the card room were ploughed back into the club for the benefit of the members. The appellant had no licence which would have been needed under Part I of the Act, if gaming had been organised at the club in these circumstances. He contended that he was not engaged in organising gaming as TH Poker was not the playing of a game of chance within the meaning of the Act. Following a complaint from the Gaming Board, police made covert visits to the club in December 2004 and January 2005.

The appellant was charged under s. 8 with an offence of organising gaming in which a levy was made in December 2004 and organising gaming where a fee was charged to participate in January 2005. He was tried before HH Judge Wilkinson and a jury at Snaresbrook Crown Court in January 2007. The appellant accepted that he had organised the game of TH Poker, a charge had been made to players for participating in the game and a levy had been made on the winnings; in the circumstance, if the prosecution proved that playing TH Poker was gaming, then he accepted the offences had been committed. The sole issue was therefore whether TH Poker was a game of chance.

It was common ground that the successful playing of TH poker required considerable skill. The appellant called expert evidence from Professor Kelly and Nick Szeremeta. No expert evidence was called by the prosecution.

It was the evidence of Professor Kelly that TH Poker was a game of chance and skill. It required more skill than any other game of poker. Even though the first two cards drawn by each player were completely at random and subsequent cards also drawn at random, the element of skill was predominant. For example, each player might get to know 5/7^ths of the hand of each other player when the final card was drawn; meaningful predictions could then be made by the skilled player. He did not commit to a ratio of skill to chance; research work had shown that the ratio was 70:30 of skill to chance. He accepted that a 30% chance was not insignificant.

Mr Szeremeta gave evidence that some types of poker required more skill than bridge; skill was more significant than chance. He gave detailed evidence about the mathematical calculations necessary to explain the odds and the strategies employed.

On 16 January 2007, the judge heard argument as to the directions to be given to the jury. The appellant contended (on the basis which we will set out) that the test as to whether TH Poker was a game of chance depended on whether skill predominated over chance; as TH Poker was predominantly a game of skill, it was not a game of chance within the meaning of s. 52(1). The judge in a short ruling rejected that submission and held that no gloss was required on the definition in the Act, though he accepted the prosecution submission that the prosecution had to prove that there must be a significant element of chance, though not necessarily a predominant one, as most games contained some elements of chance even to an infinitesimal extent.

The judge then summed up the law on what constituted a game of chance to the jury on the basis of his ruling:

The appellant was convicted on 16 January 2007. He was subsequently conditionally discharged for two years and ordered to pay £10,000 towards the prosecution costs.

He appeals against his conviction to this court by leave of the single judge.

The contention elegantly and succinctly advanced by Mr Luba QC on behalf of the appellant was that Parliament had radically altered the law in 1960, that it was no longer relevant to examine the old cases and that on the true construction of the 1968 Act, the judge should have directed the jury that a game where skill predominated over chance was not a game of chance within s.52(1) of the 1968 Act.

Mr Luba QC’s argument was founded on a succinct, clear and very helpful historical analysis of the development of the law. This can be summarised as follows:

At common law, the playing of any game of chance, including dice and card games, was not unlawful: The Case of Monopolies (1603) 11 Co Rep 87; Sherbon v Colebach (1691) 2 Vent 175. However, the keeping of a common gaming house was punishable at common law as a nuisance: Hawkins Pleas of the Crown (book 1 c.75, s.6) made clear that:

The prohibitions on games of chance and games of skill were all statutory. From an early time, Parliament had placed restrictions on playing games of chance. In 1541, by the Unlawful Games Act (33 Hen 8 c.9), Parliament had prohibited certain games at certain locations at certain times. Various enactments changed the law over the centuries; some restricted the circumstances in which games might be played; others prohibited certain games. The Gaming Act 1845 (8&9 Vict c 109) made significant changes; it legalised all games of skill, but preserved all the penalties which attached to the playing of unlawful games anywhere, or gaming at all (even of lawful games) in common gaming houses. It made specific provision for proof of what amounted to a common gaming house:

In Jenks v Turpin (1884) 13 QBD 505, the Divisional Court considered a case stated on the issue as to whether a person who ran a club where baccarat was played and entrance fees charged was keeping premises for unlawful gaming contrary to s.4 of the Gaming Houses Act 1854. The court decided that the premises were kept for unlawful gaming if the place where the activity occurred was a common gaming house or if the game was illegal; on the facts the club was a common gaming house and baccarat was an illegal game.

In reaching the conclusion that Baccarat was an unlawful game Hawkins J (at page 523) considered that unlawful games could be divided in two classes:

The test set out by Hawkins J was consistently followed in a series of cases in the years down to 1960; it is only necessary to refer to three:

In Pessers, Moody, Wraith and Gurr Ltd v Catt (1913) 77 JP 429, suppliers of a machine through which a game was played sued the defendant who had rented it for royalty payments. The defence was that the machine could not be used as the game was a game of chance and not a game of skill and therefore unlawful. All the judges who considered the issue Eve J (on a motion for an interlocutory injunction), Scrutton J (at first instance), Vaughan Williams LJ, Farwell LJ and Kennedy LJ (in the Court of Appeal) considered that there was more than a scintilla of skill and it was a game of skill; Kennedy LJ gave the fullest reasons:

In Dalton v Adelphi Club [1938] 4 All ER 556, the club which organised games of stud poker on their premises were prosecuted under the Gaming Houses Act 1854. The metropolitan stipendiary magistrate found it was a game of skill. The Divisional Court presided over by Lord Hewart CJ held that it was not a game of mere skill; whatever degree of skill an experienced player might acquire, the game of stud poker as described in the case stated was always a game of chance; it could not be described as a game of mere skill.

In R v Tompson [1943] 1 KB 650, the defendants who ran a bridge and poker club on the Finchley Road in London were indicted under the 1854 Act at the County of London Sessions. Expert evidence was called to show that the way poker was played at the club was a game where skill predominated over chance. The chairman ruled that the question as to whether poker as carried on at the club was an unlawful game was a question for him and he held it was, as it was not a game of mere skill. The Court of Appeal presided over by Viscount Caldecote CJ held he was wrong; the court said at p 656:

The court pointed out that the cases dealing with machines were different; since 1913, when Kennedy LJ formulated the test set out above,

Mr Luba QC submitted that the change in the law effected by the 1960 Act had the consequence that it was no longer appropriate to rely on the older cases to which he had referred us. The Act effected a fundamental change, as the short summary which we have set out demonstrates; the Act rendered gaming lawful on the conditions set out in the Act and repealed the old statutes. The complexity of the old law with different tests applicable to cards and automatic machines was no longer relevant. The test as to gaming and a game of chance was set out in the Act and it was for the jury to determine whether on the facts the game was a game of chance as defined in the Act.

We agree with that submission. As Parliament had made a significant change of the applicable law and provided a statutory definition of a game of chance, it is, in our view, no longer necessary or helpful to refer to the old cases. In the present case, reference was made to R v Tompson before the judge; we can quite understand why, in the absence of any authority on the post 1960 regime, it was done. However, Mr Luba QC was correct in saying that it was wrong to do so.

In relation to the definition in the 1968 Act to which the judge should have sole regard, Mr Luba QC submitted that it was clear that the definition in the Act could not be applied literally; construed properly, he contended that it was clear that a game which was predominantly a game of skill was not a game of chance within the statutory definition.

First, he relied on an observation in Armstrong v DPP [1965] AC 1262. The defendant in that case ran a postal bingo club; he was prosecuted for running a lottery contrary to the Betting and Lotteries Act 1934. It was admitted that there was a lottery, but it was contended that the defendant could rely on a provision of the Gaming Act 1960 which provided that nothing in the 1934 Act made unlawful any gaming conducted in such a way that no offence was committed under the relevant part of the 1960 Act; in connection with that issue a question of law arose as to the element of participation required to constitute “playing a game” of chance within the meaning of s. 52(1) of the 1960 Act. Lord Pearson (who gave the only substantive speech) considered that on the facts there was no participation by those who played bingo as could reasonably be said to constitute the playing of a game. At page 1280D, he observed that the Act contained only the partial definition of a game of chance. Mr Luba QC relied on this very short observation for the submission that it was necessary to provide a fuller definition of the game of chance than that provided in the Act. We do not think that this provides any real assistance as the particular observation was directed at what was meant by “playing” a game of chance and not whether a particular game constituted a game of chance – see the argument of counsel for the appellant at page 1276E.

Mr Luba QC next submitted that the primary object of the definition was to ensure that any game of chance which included an element of skill was nonetheless to be treated as a game of chance. There was, however, nothing to suggest that Parliament intended games which were predominantly games of skill to be considered games of chance. Indeed as every game included an element of chance, construction of the 1968 Act which led to the inclusion within the definition of “games of chance” games such as chess and scrabble would be absurd; it could not therefore be right, as suggested in Halsbury Laws of England, Vol 4(1) 4^th edition para 3 that only games of pure skill where there is no element of chance, are excluded from the definition.

In support of this submission Mr Luba QC referred us to a number of cases decided in the USA where the courts had considered what was meant by a “game of chance”. A considerable number of decisions, from courts in Massachusetts, Ohio, North Carolina, California and Alaska were put before us in a bundle of authorities; specific reliance was placed on the summaries of the law by Judge Dull in Stubbs v Dick (1949) 89 N.E.2d 480 (Court of Common Pleas of Mercer County Ohio), by the Supreme Court of North Carolina in State v Stroupe (1953) 76 S.E. 2d 313 and by the Supreme Court of California in Re Allen (1962) 377 P. 2d 280. In those cases the courts referred to well established rules:

Mr Luba QC also relied on the minority judgment in the decision of the Supreme Court of Canada in Ross, Banks and Dyson v The Queen (1968) 70 DLR (2d) 606. The Supreme Court had to consider whether the game of contract bridge fell within a statutory provision which stated “ “game” means a game of chance or mixed chance and skill”. The majority judgment (delivered by Pigeon J) held that the provision in question was clear:

The judge had not therefore applied the test in the 1968 Act, but had based his summing up on the old law; he should have directed the jury that a game in which skill predominated over chance, was not a game of chance. There had been a misdirection.

Although, for the reasons we have given we agree with Mr Luba QC that it was not appropriate to have regard to the old law and that the issue is one of construction of the provisions of the 1968 Act, we consider that the judge directed the jury correctly on the 1968 Act:

The meaning of a game of chance set out in s.52(1) of the 1968 Act is not by its terms an exhaustive definition, as the word “include” is used. However, it does not seem to us that this was intended by Parliament to enable a more restrictive definition to be given. It is clear that Parliament could have adopted a test of preponderance; it did not and we see no reason to write into the Act a further restriction or qualification which Parliament could easily have included but which it did not.

In our view, the definition in the Act is in simple terms and needed little elaboration; it was a question of fact for the jury to determine whether on the statutory definition TH Poker was a game of chance.

It may be in some cases the definition would need some elaboration. If a prosecution was brought where the element of chance was insignificant or de minimis, then it would be necessary to spell out that that element of chance should be ignored in determining whether the game is a game of chance. For example, if chance was be used to determine which player had the right to start a game, but the game was otherwise a game of skill, then that element should be regarded as insignificant or de minimis and therefore should be ignored.

In his direction to the jury the judge may have gone further than this in favour of the appellant. He directed the jury that there must be a significant or meaningful element of chance as opposed to an element which was simply token, notional or a scintilla. In our view, as Parliament has provided that games of combined skill and chance are to be treated as games of chance without any qualification, then the only circumstance where chance should not be taken to make a game of skill and chance a game of chance is where the element of chance is such that it should on ordinary principles be ignored – that is to say where it is so insignificant as not to matter. Parliament did not provide that in a game of mixed skill and chance that the element of chance had to be significant for the game to be a game of chance; there is no reason for the courts to do so.

It seems to us that the element of absurdity to which so much weight was attached on behalf of the appellant is properly catered for by ignoring chance where the element of chance is so insignificant as not to matter.

We have reached this conclusion on the basis of our interpretation of 52(1) of the 1968 Act. It was common ground that s. 52(6) was directed at the operation of games played against “the bank” such as in casinos or where gaming machines are used, as it refers to games “played otherwise than against one or more other players.” It was argued on behalf of the appellant that, as Parliament had considered it necessary to refer to “superlative skill eliminating the element of chance” in relation to games played against “the bank”, Parliament had envisaged courts would, in determining whether a game of combined skill and chance was a game of chance, have regard to the predominance of the elements of skill and chance. We do not consider that this in any way follows; on the contrary, on the definition of a game of chance as set out in the Act, if there was no element of chance (as that had been eliminated by superlative skill), it would not be a game of chance and skill combined. The subsection was therefore directed at bringing within the definition games against the bank where (if such exist) skill had eliminated any element of chance.

Argument was also directed to the question of whether assistance as to the statutory definition of a game of chance could be derived from s. 40(2) which provided an exemption from s.3 in certain circumstances in respect of miners’ welfare institutes or clubs. It was contended by the Crown that the exemption was directed in part at bridge clubs and, if the test was one of predominance of chance over skill, then that provision would have been unnecessary. We do not think that this subsection really assists either way.

We were grateful to Mr Luba QC for referring us to the US and Canadian authorities. We do not think that any real help can be derived from the US authorities where the concept of predominance has become embedded. Although the definition of game in the Canadian statute under consideration was exhaustive, the approach of the majority of the Supreme Court of Canada in Ross to the construction of the Canadian legislation is very much the approach we have adopted in relation to the UK legislation; for the reasons that are evident from this judgment, we were not persuaded by the views of Spence J.

There was in our view therefore no misdirection in the summing up. The jury were on the evidence plainly entitled to conclude that the game of TH Poker was a game of chance as defined by the 1968 Act. The appeal must be dismissed.