In a much-publicised recent case, the Supreme Court has considered two issues: first, whether it is necessary to prove dishonesty in order to make out an offence of cheating under the Gambling Act 2005; and second, what the test for dishonesty should be.

Mr Ivey is a professional gambler. He played a number of games of Punto Banco at Crockfords over two days in August 2012, with the help of another professional gambler, Ms Sun. Punto Banco is played with six or eight packs of cards. It requires the dealer to deal two or three cards, face down, onto two positions on the table ("punto" and "banco"). The gambler places a bet on one of those positions, and if the total of the cards dealt is closer to nine than the other position (subject to the rules of the game), he or she wins.

There is an advantage to the gambler in knowing which cards are "high value" which, in this context, means cards with a face value of seven, eight or nine. That would not, of course, normally be possible. However, Mr Ivey sought to take advantage of a technique called "edge-sorting". This can be done where the manufacturing process means that the pattern on the back of the card is marginally closer to one long edge of the card than the other. The Supreme Court described the difference as "sub-millimetric". The difficulty lay in ensuring that the cards were sorted such that one type of long edge appeared for the high value cards, and not for the others. This was Ms Sun's role. As the cards were turned face up at the end of each coup, she indicated to the croupier which cards were "good" (asking her to turn them one way) and which were "not good", asking her to turn them the other (the croupier believing that Ms Sun was simply superstitious). The use of a shuffling machine (at the request of Mr Ivey) meant that the cards were not rotated when they were shuffled and Mr Ivey also asked to keep using the same cards. By the time the sorting process was finished, Mr Ivey's bets per coup increased. By the time he stopped playing, his success rate had risen markedly. His bets for the last three coups averaged £150,000 each time, and he ultimately won in excess of £7.7 million. Alarmed by the size of its loss, Crockfords reviewed its footage of the game, and the cards, and worked out how Mr Ivey had been so successful. It therefore refused to pay his winnings.

The parties agreed that there was an implied term in the contract between Mr Ivey and Crockfords that he would not cheat. There is also, as indicated above, a statutory offence of cheating. The Supreme Court held that cheating meant the same thing in either case.

Mr Ivey admitted edge sorting, but was adamant that this did not amount to cheating. It was said on his behalf that cheating necessarily involves dishonesty. In order for him to be held to be dishonest, the relevant legal test (in relation to the criminal offence) required Mr Ivey to have known that his conduct (viewed objectively) was dishonest. He did not see it as dishonest. He had therefore not cheated, and should recover his winnings.

The Supreme Court therefore considered two issues: (i) whether there was any requirement to show dishonesty; and (ii) if so, whether Mr Ivey was dishonest applying the proper test.

In relation to the first issue, the Supreme Court held that it is not necessary to show dishonesty, in order to prove that someone has cheated. Lord Hughes accepted that the concept of honest cheating is an improbable one. It is not, however, impossible. Lord Hughes provided several examples, such as tripping an opponent in a race, giving a horse too much water before it is due to run, or deliberate time-wasting in a number of sports. The Supreme Court emphasised the role of dishonesty in some cases as supplying the necessary element of "illegitimacy and wrongfulness", but that, in the case of cheating for example, the cheating itself carried its own inherent stamp of wrongfulness.

The Supreme Court went on to consider what the proper test for dishonesty was. In relevant criminal cases, judges have been required for the last 35 years to direct the jury to consider the so-called Ghosh test.8 The jury has been directed to consider dishonesty in two stages: (i) was the conduct complained of dishonest by the lay objective standards of ordinary, reasonable and honest people; and (ii) if so, whether the defendant must have realised that ordinary honest people would so regard his or her behaviour. The Ghosh test is therefore usually described as involving both an objective and a subjective test.

Lord Hughes, giving the judgment of the Supreme Court, identified six problems with the second, subjective limb of the test. These included the unintended effect that the more warped the defendant's standards of honesty are, the more likely he or she is to be acquitted, and the "unprincipled divergence between the test for dishonesty in criminal proceedings and the test of the same concept when it arises in the context of a civil action". He further held that the subjective limb of the test had been introduced on the basis of a misunderstanding of earlier authorities.

The Supreme Court therefore came to the conclusion that the subjective limb of the Ghosh test does not correctly represent the law. Going forward, in both civil and criminal cases, the test will be that currently used in civil cases – the judge or jury (depending on the type of case) will first have to ascertain the actual state of the individual's knowledge or belief as to the facts. Once that is established, the question of whether his or her conduct was honest or dishonest is to be determined by applying the standards of ordinary decent people. The test therefore retains a role for the state of mind of the individual defendant, but it removes the requirement for him or her to appreciate the dishonesty of his or her actions.