Calvert v William Hill Credit Ltd

This case raises, for the first time in an English court, the question whether there are any circumstances in which a bookmaker can incur liability in negligence in respect of the gambling losses of a customer who is, and who is known by the bookmaker to be, a problem gambler. More specifically, the question is whether a bookmaker who has, at the customer’s request, undertaken to prohibit the customer from gambling for a specified period, owes the customer a duty to take reasonable care to enforce that prohibition, so as to protect the problem gambler from the risk of gambling losses during the specified period.

In Reeves v. Commissioner of Police [2000] 1 AC 360, at 368, Lord Hoffmann said that:

“…there is a difference between protecting people against harm caused to them by third parties and protecting them against harm which they inflict upon themselves. It reflects the individualist philosophy of the common law. People of full age and sound understanding must look after themselves and take responsibility for their actions. This philosophy expresses itself in the fact that duties to safeguard from harm deliberately caused by others are unusual and a duty to protect a person of full understanding from causing harm to himself is very rare indeed….”

Lord Hoffmann was speaking about harm in the form of personal injury or death. Although the unrestrained continuation of gambling, whether profitable or loss-making, may cause or aggravate what is now a recognised psychiatric disorder, the harm for which compensation is mainly sought in the present case, namely the financial ruin caused by gambling losses, represents economic loss. It follows that the recognition of a common law duty to protect a problem gambler from self-inflicted gambling losses involves a journey to the outermost reaches of the tort of negligence, to the realm of the truly exceptional.

The question whether there existed a duty of care in the present case is further complicated by the fact that the relevant gambling losses were sustained during the period of gestation of a radical change in the attitude of the law towards gambling, which was itself the product of changing social attitudes. Although the common law did not impose any general disability on gaming contracts, section 18 of the Gaming Act 1845 provided that all gaming or wagering contracts should be null and void. The policy behind this legislation was to discourage gambling: see Hill v William Hill (Park Lane) Limited [1949] AC 530, at 548 per Viscount Simon.

In its March 2002 White Paper entitled “Safe Bet for Success”, proposing a new legislative and regulatory framework for the gambling industry, the Department for Culture, Media and Sport said this, under the heading “Dealing with the Downside”:

“In the Government’s view the law should no longer incorporate or reflect any assumption that gambling is an activity which is objectionable and which people should have no encouragement to pursue. It is an important industry in its own right, meeting the legitimate desires of many millions of people and providing many thousands of jobs.

But gambling also presents particular risks to children and the vulnerable which other forms of leisure do not. Too early exposure to gambling can be harmful; and for some people the temptation to gamble to excess is very hard or in practice impossible for them to control. While the law should be morally neutral to gambling, it should, as the review body recommended, also provide proper controls and protections for those who may be or already have been damaged. By international standards the incidence of problem gambling in Great Britain seems to be low. But there are no grounds for complacency. There is not yet a reliable run of figures; and even on the low rate of problem gambling suggested by a recent survey there are still between 275,000 and 370,000 problem gamblers at any time.

It is impossible to do away with problem gambling; and excessive controls could make matters worse by encouraging the growth of illegal gambling. The Government does not think that, at least for the time being, it would be sensible to try to put in place a numerical target for reducing problem gambling. But it is clear that the law should provide assurance that all parts of the industry will operate to the highest standards of social responsibility, recognising, that the strength of the controls embodied in the law will need to be kept under careful review and adjusted if necessary.

There are therefore two sides to the issue: a set of statutory safeguards governing specific gambling activities, and – running alongside them – a commitment by all licensed gambling operators to conduct their business in a way which is socially responsible. The Government agrees with the Review Body’s conclusion that the Gambling Commission should issue formal codes of practice in relation to social responsibility which should become part of the conditions of licences to operate. These codes should cover such matters as the avoidance of encouragement of children to gamble; provision for players to bar themselves from gambling; the display of clear information about the probabilities of winning and losing; and the provision of information to customers about problem gambling and what people who think they might need help should do. The codes should apply as much to gambling provided on the internet or through interactive television as to traditional gambling outlets.”

That summary of policy was in due course reflected in the Gambling Act 2005. With effect from 1^st September 2007 the longstanding statutory prohibition on the enforcement of gaming contracts was repealed. More generally, the Act established a new licensing regime supervised by a new Gambling Commission. Section 1(c) of the Act defined as one of the three licensing objectives “protecting children and other vulnerable persons from being harmed or exploited by gambling”. Section 22 required the Gambling Commission to “aim … to pursue, and wherever appropriate to have regard to, the licensing objectives”, and for that purpose to prepare, and keep under review, a Statement of Principles.

The quid pro quo for the removal of the traditional ban on the enforcement of gaming contracts was that persons providing facilities for gambling activity were subjected to a licensing requirement which might (and in due course did) impose conditions including the putting into effect by licensees of policies and procedures intended to promote socially responsible gambling, including a commitment to the identification and treatment of problem gamblers. The Commission was by section 24 of the Act required to issue codes of practice about the provision of gambling facilities describing (in particular) facilities for the purposes of “making assistance available to persons who are or may be affected by problems relating to gambling”. By section 24(8) a failure to comply with a provision of a code did not of itself give rise to criminal or civil liability but, by section 24(9) a code:

“(a)

shall be admissible in evidence in criminal or civil proceedings,

(b)

shall be taken into account by a court or tribunal in any case in which it appears to the court or tribunal to be relevant,

…”

The Gambling Commission published its Licence Conditions and Codes of Practice in November 2006 and June 2007. Section 2.1 of the Commission’s principal Code of Practice required licensees’ policies and procedures for socially responsible gambling to include certain specific policies and procedures designed to combat problem gambling including, under section 2.5, headed “Self-Exclusion”, the following requirement:

“Licensees must put in place procedures for self-exclusion and take all reasonable steps to refuse service or to otherwise prevent an individual who has entered a self-exclusion agreement from participating in gambling.”

The participants in the United Kingdom gambling industry, of which the William Hill group is a prominent member, did not wait for the coming into force of the Gambling Act 2005, or for the publication of the Commission’s Licence Conditions and Codes of Practice before taking active steps to establish their own policies and procedures, for the purpose of demonstrating and putting into effect a commitment to social responsibility. For present purposes, it is sufficient to refer to the steps taken by the Association of British Bookmakers (“the ABB”) and the Remote Gambling Association (“the RGA”). The William Hill Group is a member of both of them. The ABB (founded in 2002) is an association of Licensed Betting Shop operators. It issued its Social Responsibility and Good Practice Code for Licensed Betting Offices in association with GamCare, a charitable body set up and funded by the gambling industry for the provision of assistance to problem gamblers. It was published by 2006, and contains in heavy type the following introduction:

“The responsibility for an individual’s gambling is their own; the responsibility to exercise a duty of care is that of the operator.”

Under the heading “Commitment”, paragraph 1 continued:

“Apart from the fact that a responsible gambling company should adopt a duty of care, changing legislation will to some extent compel this to happen.”

Paragraph 7 headed Problem Gamblers reads (in part) as follows:

“Dealing with a suspected problem gambler is a delicate matter that needs to be approached sensitively. An appropriate response should include:

•

Rules and Procedures regarding problem gamblers should be written and incorporated into the day-to-day operation of the LBO.

•

All staff should be trained to deal with situations that might require help and advice.

•

A positive response, including communicating sources of help and the possibility of self-exclusion to any customer who admits to having lost control of their gambling or who is exhibiting symptoms of acute distress which may be caused by their gambling.

•

A ‘self-exclusion’ policy, supported by written instructions for staff, for any customer who wishes to be barred from LBOs operated by the operator.

•

A minimum self-exclusion period of six months in all circumstances.

•

A written statement accepting terms of exclusion, that is signed and retained by the customer as well as by a staff member of the LBO (and supported by a photograph of the customer if possible).”

The following relevant material appears on the ABB website. (I have not been able to ascertain the date of its first appearance.) Under the heading “ABB, Bookmakers and Social Responsibility” there is a statement that:

“They have a duty of care to punters and should work to minimise the incidence of problem gambling.”

The website includes a sample customer self-exclusion agreement which is designed to identify the betting shops from which the problem gambler is intended to be excluded, and which continues as follows:

“If I attend any of the premises set out in this agreement during the term of this exclusion, and I am identified by [name of bookmaker], I will be requested to leave the premises. If I refuse and/or become a trespasser I will be removed.

I release [name of bookmaker], its manager(s) and employees from any liability or claims in the event that I fail to comply with this voluntary exclusion.”

As its name suggests, the RGA is an association of businesses providing facilities for gambling otherwise than face to face (as between customer and bookmaker), and in particular for gambling on the internet and by telephone. It had by 2006 published a Social Responsibility Code with, as one of its two stated objectives, that of encouraging social responsibility within the betting and gaming industry, to be effected through various means including support for charities and initiatives to help those who have gambling problems. The Code appears to have been designed primarily in relation to internet gambling but at paragraph 3 it is stated that:

“The underlying principles hold good for all forms of remote gambling.”

Under the heading Player Protection Measures, paragraph 8 includes a number of suggested measures whereby operators can help customers, for the purpose of preventing and combating problem gambling. The final one is as follows:

“8 …

•

The provision of a self-exclusion facility for any customers who wish to exclude themselves from gambling on the operator’s website. During the chosen period the customer will also be excluded from all forms of communication from the operator. The chosen period should be for a minimum of six months.”

Licensed Betting Offices run by the William Hill Group are and have for many years been a familiar site on high streets in the United Kingdom. Under the auspices of William Hill Organisation Limited, a subsidiary of William Hill Plc (the Group parent company) the betting offices constitute the Group’s retail gambling division. There are in excess of 2,000 William Hill licensed betting offices in the United Kingdom, Channel Islands, Isle of Man and Ireland.

Less conspicuous to the general public is William Hill’s remote gambling division, operated through the defendant William Hill Credit Limited, another subsidiary of William Hill Plc. Although it provides internet and telephone gambling services, the present case is concerned with its telephone gambling service. This was operated from two linked call centres, in Leeds and Sheffield, staffed by some six hundred employees, serving the telephone betting activities of some 180,000 customers and for that purpose receiving some 20,000 calls per day. Incoming calls were handled in the first instance by teams of telephonists, supervised by Team Leaders.

Risk management within the William Hill Group was, in relation to all its betting activities (shops, telephone and internet), handled by a single Betting Control Department, which was sub-divided into two teams, Odds Compilers and Risk Managers. The former were charged with setting and adjusting the odds offered by William Hill on any sporting event. The Risk Managers were charged with monitoring and regulating the taking of bets from customers. During 2006 the Risk Management team consisted of eighteen persons. Generally speaking, telephonists had authority to accept bets up to certain levels, varying according to the nature of the event and the identity of the customer, beyond which they were required to seek authorisation from a Risk Manager, either by internal telephone or orally from a duty Risk Manager known as a “Floor Walker”, who would patrol the call centre within sight of the telephonists.

The Telephone Betting Division maintained a dedicated team for premium customers known as the Golden Spurs Team. Premium customers consisted of sport professionals such as trainers, members of the press, celebrities and other high value customers. The Golden Spurs Team worked in close cooperation with and proximity to the Risk Managers, and its customers were encouraged to use a dedicated telephone number, so as to streamline the monitoring and (if thought fit) timely acceptance of their bets.

The operations team of telephonists and Team Leaders were assisted in dealing with William Hill’s telephone betting customers by its Customer Services Department. Use was also made of an age verification service provided by an outside contractor, to minimise the risk of dealing with under-age gamblers.

The central management tool with which the telephone betting division handled its customers’ bets was the customer account. These included credit accounts, debit accounts and cash accounts. For present purposes, only the second and third of those types need be considered in any detail. A customer using a debit account funded his gambling by making money transfers by credit card or bank debit card. Winnings were credited to the account and net balances paid from time to time to a bank account nominated by the customer. There was, in 2006, no limit (whether daily or otherwise) to the amount which a customer could deposit in a debit account, nor any general limit on the amount which he could bet, either on a single event or during a day, week or month. In practice however, as will appear, the Risk Managers exercised close control over the acceptance or rejection of large bets, using computer systems which enabled them to monitor both William Hill’s moment to moment exposure in relation to relevant sporting events and the precise betting history of particular customers.

The cash account differed from the debit account only in the manner by which it was funded. As its name implies, a customer with a cash account would fund it by delivering cash to his nearest William Hill betting office, rather than by credit or debit card payment. As will appear, it was possible for a single customer to have, at one and the same time, more than one type of account with William Hill, and even more than one account of the same type. I was told that, in 2006, new accounts were being opened with William Hill’s telephone betting division at the rate of some 70,000 per annum.

The procedure for opening a telephone betting account with William Hill was in 2006 (and remains) as follows. The standard response by a telephonist to an incoming call was to ask the caller for his account number and name. A customer with no existing account would then be asked for his credit or debit card details, his forename and surname, address and contact telephone number, and then his date of birth. The telephonist would then ask whether he wished to have mail shots, promotions and betting statements sent to his address (to which request I was told that up to 40% of customers said no). Meanwhile, the telephonist would by inputting the information received from the customer into her computer terminal be able to generate a telephone betting account number, and provide it orally to the customer.

In the case of a debit account, the customer would be able immediately to fund the account by a credit or debit account transfer and commence betting there and then. It would however take William Hill up to 72 hours to obtain independent age verification, prior to which no payment of net winnings would be made to the customer.

A feature of the William Hill debit account system was that each account could be funded only by a single credit or debit card. A customer wishing to use more than one card to fund his gambling would need to open a new account for each additional card. Furthermore, a customer whose card had expired would need either to revalidate his existing account using his new card, or open a new account.

The procedure for opening new accounts by an existing customer was substantially the same as that which I have described in relation to a new customer save that, if he could recall his existing account number and provide it to the telephonist, the process of obtaining his personal details would be short circuited by the telephonist being enabled to have access to the details on the existing account, and no further age verification check would be required. By contrast, where a customer either could not or would not provide an existing account number or linked bank card number, the telephonists had no means at their disposal of obtaining access to the customer’s existing account details. The account number and card number were the telephonist’s only keys into the account. It was thus not merely possible but routine for a telephonist to open a new account for an existing customer without being aware of, or having access to that customer’s account details. Unless the customer chose to identify himself as such, the telephonist would not be aware that the caller was an existing customer at all.

By contrast with telephonists, Team Leaders, Risk Managers and Customer Services staff could obtain access to existing customers’ accounts without being provided with account numbers, for example by a name search, refined if necessary by reference to postcode, detailed address or by a check of date of birth, where the other details identified more than one person of the same name at the same address, such as a father and son. Since however the mere opening of an account was a routine and often repeated task, it was not commonly referred up by telephonists to a Team Leader, and betting on a newly opened account would, if necessitated by the size or frequency of bets, be referred by the telephonist to a Risk Manager rather than to a Team Leader.

The final stage in the creation of an account for a new customer consisted of the transmission from the telephone betting division to an outside contractor of the name and address details sufficient to enable the contractor to send the new customer a welcome pack. This included William Hill’s Telephone Betting Rule Book setting out its terms and conditions of betting. They were also available on the William Hill website. The account opening script used by telephonists required the new customer to be invited to accept William Hill’s standard terms and conditions, referred to them being available on the website and in the telephone rule book, and the customer would receive his account number only upon confirming acceptance. I consider it likely that very few telephone betting customers read those terms and conditions before giving that confirmation, obtaining their account number and starting to bet.

William Hill implemented its own response to the need to conform with the codes of practice to which I have referred in 2004. I heard evidence from Mr William Sharkey, the head of William Hill’s telephone betting division from September 2005, who was personally involved in the formulation of William Hill’s policy, and the carrying out of the necessary staff training. Apart from a reluctance to make what he thought might be damaging admissions, he was an honest and reliable witness. The 2004 policy remained in force during the period in 2006 material for these proceedings, but has since been revised. For present purposes, the most important record of the policy appears in a document entitled “Procedures for providing information on responsible gambling, customer self-exclusion and if notified a customer is under eighteen” in the version aimed at internet and telephone gambling. In the summary which follows, I shall omit reference to the under-age gambling provisions.

William Hill’s self-exclusion procedure was to be triggered by any of the following three events:

If any of those events occurs, the telephonists’ responsibility is simply to refer the customer up to their Team Leader. A reference to the Team Leader of the first type (problem gambling statement by customer or relative) requires the Team Leader to refer the matter immediately to a Customer Services representative. A reference of the second or third types (account closure or self-exclusion request) requires the Team Leader to ask whether the reason for closing the account is that the customer is experiencing difficulty controlling their gambling, and if so to pass the matter to the Customer Services representative.

The Customer Services representative is charged with three tasks:

The responsible gambling leaflet directs a customer both to GamCare and to the possibility of self-exclusion. The GamCare Helpline enables the customer to obtain skilled advice on his gambling problem. Step 3 requires closer analysis, and is best summarised by identifying its objectives.

The first is to explain to the customer the distinction between self-excluding, designed to prohibit the customer from reopening his account for an initial six month period, and mere account closure, which will leave the customer free to reopen his account at any time. The second is to explain to the customer the main provisions of the Self-Exclusion Agreement, and to enable the Customer Services representative to complete it by filling in the relevant details. Those include the customer’s name, internet and telephone gambling account numbers, and the numbers of any additional credit and debit cards (beyond those linked to the identified accounts) which a customer wishes to disable himself from using for betting. The Customer Services representative is required to emphasise to the customer that self-exclusion operates by reference to identified account numbers rather than by reference to the customer’s name.

The main terms of the Self-exclusion Agreement include the following:

“At the end of this period of six months, he/she can review this exclusion with the Customer Services Manager and either renew it for another six months or decide that further self-exclusion is unnecessary. If the customer attempts to use or re-open the above accounts during the term of exclusion he/she will be refused permission to do so. The customer recognises that it is his/her primary responsibility to identify and notify William Hill of all of his/her accounts.

The customer agrees that he/she will not place any bets, or enter into any gaming activity, with the William Hill Group’s Internet or telephone betting or gaming operations during the period of self-exclusion. If he/she does try to place a bet or enter into any gaming activity and in the unlikely event that the bet is accepted or the customer is allowed to enter into the gaming activity, he/she will not make a claim for the return of any stakes placed or otherwise once the event has been completed.

The customer releases all companies from within the William Hill Group, their officers and employees from any liability or claims whatsoever in the event that they fail to comply with this voluntary exclusion.”

The next objective is that the customer should be warned that if he wishes to exclude himself from betting at a William Hill betting office, he should visit that shop directly and inform the manager on site, if necessary providing a photograph to assist in identification. No separate communication was required to be, or was, made by the Customer Services Department to any part of the William Hill retail division.

The final objective was to achieve the closure of the customer’s telephone betting (and if relevant internet) accounts. In relation to telephone betting accounts, this was achieved by placing a “Vet Code 91” notice on all open accounts, together with a note that the account could not be reopened for six months, and by the recording on those accounts of any additional credit or debit card numbers supplied by the customer but not already linked to them. By contrast, a customer who, after advice, elected merely to close his account rather than to self-exclude, would have his account closed by a “Vet Code 92” endorsement.

The written procedures required copies of completed self-exclusion agreements both to be retained by the Customers Services department and sent to the customer, for all correspondence with the customer to be retained, for all oral communications with the customer to be recorded on tapes, and for those tapes also to be retained. Finally, the Customer Services Department was required to report all self-exclusion agreements to the relevant telephone and internet operating departments. Mr Sharkey initially told me that in 2006 he was notified of telephone betting self-exclusion agreements at the rate of approximately 20 per month. He made routine monthly reports on self-exclusions to his superiors. In further cross-examination of Mr Sharkey after late disclosure of further documents it appeared that the figures in 2006 may have been significantly lower. On the basis of a six month initial period of self-exclusion, it follows that there are likely to have been, at any one time in 2006, no more than approximately 120 self-excluded betting customers, and probably significantly less.

The procedures which I have described were buttressed by detailed scripts to be used by telephonists, Team Leaders and Customer Services representatives, and I was shown staff training materials used both in 2004 and thereafter in relation to newly recruited staff, designed to introduce and explain those procedures. It was not clear either from those materials or from Mr Sharkey’s evidence precisely to what level each of those different types of staff were trained, but no case was advanced to me that there was any structural weakness in the process of staff training. Nonetheless, as William Hill’s own training materials made clear:

“The integrity of the Policy relies on staff exercising their responsibilities and, in the case of self-exclusion, supporting the wishes of the customer to be excluded.”

In particular, the policy depended critically upon self-exclusion requests, and account closure requests related to problem gambling, being passed by Team Leaders to the Customer Services Department. The training procedures contained an express requirement, in the event of a customer request made when the Customer Services Department was closed, for the request to be transmitted by the Team Leader by email to the Customer Services Department, and for the Customer Services representative to deal with that email on the following day.

In my judgment the self-exclusion policy which I have summarised, and which was in operation in its original form in 2006, contained a significant structural weakness such that, even if operated to the letter, it could fail effectively to support the wishes of the average customer to be excluded. I shall describe the typical features of problem gambling (so far as established by the medical and other evidence) in due course, but it is common ground that it consists of a vulnerability to an urge to gamble excessively which is essentially intermittent or episodic in nature. As Mr Calvert himself put it, and the experts agreed, problem gamblers have “moments of clarity” when they are aware both of the nature and extent of their problem, and of the likelihood that they will in due course succumb again to a form of tunnel vision in which all considerations of prudence, family responsibility and self-preservation will be overborne. The essential purpose of a self-exclusion policy is to enable the problem gambler to take advantage of his moment of clarity by seeking his bookmaker’s assistance so that restraint will be available when his own resources of self-control later fail. The inherent assumption is that the customer may seek to resume his gambling during the agreed period of exclusion, and may then need to be prohibited from doing so by the bookmaker.

It will be obvious from the account opening procedure which I have described in some detail that a self-excluded customer does not need to reopen an account in order to resume telephone gambling with William Hill. He may simply open a new account, using a different credit or debit card. The telephonist who answers his call will not know that he has been self-excluded and will be under no obligation to refer his account opening request to the Team Leader. The self-excluded gambler may either decline to provide or not remember the number of his closed account or linked bank card, without access to which the telephonist is in practice quite unable to access any record identifying him as a self-excluded customer.

The placing of substantial bets by a self-excluded customer who is given a new account by a telephonist acting in all innocence will rapidly come to the attention of the Risk Managers. Unlike the Operations staff, they were not in or before 2006 given any detailed training in the self-exclusion policy. They did however have full access to the accounts, including closed accounts, of customers, which they could achieve by name search. Having heard evidence from two senior managers of the Risk Management Department I am satisfied that even in 2006 the Risk Managers would probably have declined authority to accept large telephone bets from a customer who had opened a new account after previously being self-excluded, at least without making further enquiry. This is because their monitoring of the conduct of a large scale gambler would be likely to bring to their attention a Vet Code 91 notice placed on an earlier closed account.

Graham Calvert was born in June 1979. His father was (and remains) a successful greyhound trainer, based on a farm and substantial kennel at Sedgeletch Farm, Fencehouses, Houghton le Spring, Tyne and Wear, which he ran in partnership with his wife. On leaving school at 15 the claimant worked full time in his father’s business, obtaining a kennel hand’s licence almost immediately, and becoming known in greyhound racing circles as Graham Calvert Junior.

Upon his parents’ divorce in June 1999 he acquired his father’s share of the partnership business, and has since established a substantial reputation as a greyhound trainer in his own right. His father moved to Ireland and established a greyhound training business there. The claimant married in 2001 and, with money raised from a mortgage of the farm, converted a barn for use as their matrimonial home. By 2006 their family included two children, one having been born to his wife from a previous relationship. The greyhound training business owned and run by the claimant and his mother from 1999 onwards involved the breeding and racing both of their own greyhounds and of those owned by clients.

By the year 2000 the claimant had become a skilful and successful gambler. He told me that for the period 2000 to 2005 his net winnings from gambling averaged very approximately £50,000 per year. By comparison, the accounts of his and his mother’s greyhound training business suggested very modest net profits, but the claimant maintained that he and his mother also derived substantial profits from the breeding, buying and selling of greyhounds which were not recorded in the partnership’s accounts, so the claimant said, on his accountant’s advice. I have considerable doubt as to the truth of the Claimant’s explanation for that omission.

The claimant maintained in his evidence-in-chief that he had by the beginning of 2006 established himself and his family in a comfortable but not extravagant lifestyle which was not dependent upon his winnings from gambling, which he said that he pursued for pleasure. By the end of a rigorous cross-examination I became satisfied that the claimant’s lifestyle was by 2006 substantially dependent upon his gambling winnings, although I accept the gist of his evidence that he and his mother made a significant return from cash transactions connected with the sale of greyhounds which, because they were not recorded in any accounts, books or records, are now incapable of quantification.

The claimant’s gambling started “on course”, that is by placing bets with bookmakers at greyhound racing stadia. As a greyhound racing professional he was well placed to make substantial gambling profits, and until 2004 his betting was almost exclusively confined to greyhounds. During 2003 to 2004 his betting became concentrated more on betting offices than on course, and he began telephone betting in 2005.

During the period up to and including 2004 the claimant described himself as being in control of his gambling, betting mainly on events within his special expertise, and staking no more than £1,000 on any individual bet.

From 2004 onwards the claimant’s gambling began to extend beyond greyhounds, first to horses and later to a range of sporting events, including in particular football and golf. His main reason for doing so was because his growing success as a trainer of and gambler on greyhounds led bookmakers increasingly to place restrictions on his betting, sometimes by limiting the amounts of his stakes, sometimes by refusing to take bets, and sometimes by quoting him special odds. He said that the ‘buzz’ which he got from gambling was being hampered by those restrictions, but that bookmakers were less inclined to restrict him when he was gambling on events outwith his particular expertise. I accept that evidence.

There are no documentary records of the claimant’s gambling on course, and the records of his gambling at betting offices are fragmentary and therefore unsatisfactory. The result is that findings about his gambling career prior to his commencement of telephone betting in August 2005 depend entirely on the claimant’s recollection and oral evidence. By contrast, disclosure of records by all the bookmakers with whom the claimant conducted telephone betting, from August 2005 until December 2006, when he ceased gambling altogether, enabled his telephone betting activity to be ascertained in full and accurate detail. William Hill have also provided incomplete records of the claimant’s continued betting at their betting shops during the short period August to November 2006. It is common ground that this is by no means a complete picture of the claimant’s continued betting at betting offices but, once he began it, telephone betting constituted much the largest element of his betting activity.

Although it is common ground that the claimant was a problem gambler by December 2006, there are issues first when he became one and secondly whether his condition deteriorated to the point where he became a pathological gambler. I shall state my conclusions on those issues in due course, but for present purposes it is convenient briefly to summarise the claimant’s telephone gambling career up until the point when he opened an account with William Hill in May 2006. The claimant placed telephone bets first with Pagebet (previously known as Ruben Page), a modest sized bookmaker owned by a Mr Austin Carney who had known both the claimant and his father for many years. Between 15^th August and 26^th December 2005 the claimant staked a total of £238,600 on the telephone with Pagebet, from which he derived net winnings of £41,891.87. He placed bets on twenty four different days, turning in a net profit on fifteen of them. The number of bets placed on any particular day ranged generally from one to five, with a single day (13^th September) on which he placed thirteen bets. Individual stakes ranged from £500 to £15,000, with the majority lying in the range of £1,000 to £5,000.

The claimant told me that after an initial restriction of stake size, he was eventually prohibited from further telephone betting with Pagebet by Mr Carney himself, who told him that he regarded his betting as getting out of control. Whatever Mr Carney may have said, it seems to me at least as likely that Pagebet excluded the claimant on commercial grounds. To lose over £40,000 to a single customer in four months cannot have made him an attractive counterparty to a modest sized bookmaker.

Before the claimant had either been restricted or excluded by Pagebet, he had opened a telephone betting account with a larger bookmaker Stan James. During the six week period from 30^th October to 12^th December 2005 the claimant staked a total of £1,783,219 with Stan James, yielding a net return of £140,925. He placed telephone bets with Stan James on no less than twenty nine days. The number of bets placed on any single day ranged between one and eighteen. Individual stake sizes ranged between £1,000 and £50,000. He staked more than £100,000 on four different days and his highest daily aggregate stake, on 16^th November, was £343,000.

The claimant told me that he himself felt that his betting with Stan James was getting out of control, and that he accordingly closed his account in December 2005 and asked for it “never to be re-opened”. He said that he was not at that stage aware of the phrase self-exclusion. He described how, about a week later, he tried to reopen his account at Stan James but was told that there was a message on the system saying that his account was not to be reopened, and his request was therefore refused. He produced letters from Stan James written in 2007 at his request, the combined effect of which was to confirm that his account had been closed on 12^th December 2005 pursuant to his instructions that it never be reopened. Again, the claimant’s ability to make substantial profits from telephone betting with Stan James over a very short period suggests that his account might have been closed for commercial reasons. Nonetheless I am inclined to believe the claimant’s account of the circumstances in which his Stan James account was closed, corroborated as it is by Stan James’ letters.

The claimant then turned his attention to Ladbrokes, commencing telephone betting on 19^th January 2006, and placing his last bet with Ladbrokes on 20^th March. During that two month period he staked an aggregate of £1,392,619, yielding a net return of £609,311. This represented a slight reduction from the scale of his combined telephone betting in late 2005 with Pagebet and Stan James, but a substantial improvement in profitability. He placed telephone bets with Ladbrokes on twenty six different days. The largest number of bets on any single day was seventeen (on 7^th February), the largest daily aggregate stake was £186,500 (on 16^th March) and the largest single bet (a loser) was £86,666 on 5^th March. He placed aggregate bets in excess of £100,000 on four different days.

The claimant told me that prior to opening a telephone betting account with Ladbrokes, he had begun gambling in Ladbrokes’ betting offices, but that the logistics of obtaining cash from banks in sufficient amounts to enable him to place bets which generated the necessary ‘buzz’ (which sometimes required him to spend up to half a day collecting cash from a number of different bank branches) led him to prefer a telephone betting account as a much more convenient means of placing very large bets. His typical individual stakes by telephone with Ladbrokes ranged between £5,000 and £30,000.

In what he described as a “moment of clarity” in mid March 2006, the claimant said that he telephoned a Mr Mick Embleton who handled his bets at Ladbrokes and asked for his account to be closed, and not to be reopened. He said that, two weeks later, he was permitted to reopen the account after signing a statement to the effect that he was no longer a gambling addict but that he found that Ladbrokes limited his bets, for example limiting a requested bet of £50,000 on a particular event to £5,000. Accordingly, he said that he gave up using Ladbrokes any further and closed the account, requesting again that it never be reopened, before starting telephone betting with William Hill.

Mr Embleton from Ladbrokes was not called as a witness, and the claimant produced no supporting documents from Ladbrokes confirming his explanation how that account had come to be closed. In response to enquiry by the claimant’s solicitors in 2007, Ladbrokes stated in a letter merely that his account had been opened on 19^th January 2006, closed on 16^th July 2006, and that there were no self exclusion agreements in place. The entries in late March 2006 recording his final bets on the telephone with Ladbrokes do not corroborate his description of having had his bets limited. His last three bets were for £30,000, £20,000 and £10,000 respectively. Again, having made spectacular net winnings from telephone betting with Ladbrokes over a two month period, it is in my judgment equally likely that Ladbrokes limited his betting for commercial reasons. Although the claimant’s evidence is consistent with the manner in which surviving tape recordings show that he endeavoured to close his telephone betting account with William Hill on two separate occasions, both of which he was able to recall before the relevant tapes were discovered and made available, I do not accept his evidence in relation to his Ladbrokes account. Such other evidence as is available contradicts rather than corroborates his account.

There is no record of the claimant placing any telephone bets after 20^th March 2006, the date of his last bet with Ladbrokes, or during April. He said that his understanding at the time was that Ladbrokes and William Hill were the only bookmakers likely to accept very large telephone bets on a regular basis, and that he was initially reluctant to open an account with William Hill, since it had by then become the owner of the Sunderland Greyhound Stadium, where the claimant was a prominent and successful greyhound trainer. He said that he had kept his betting secret from his family, and did not wish to risk leakage from William Hill of the scale of his telephone betting via the Sunderland Stadium and his immediate circle of friends and business associates.

While I accept that the claimant did not disclose the extraordinary scale and frequency of his gambling to his family, I do not accept that he kept his gambling a complete secret from them, and I approach his explanation for an apparent six week break in his large scale telephone betting with some reserve. The evidence does not enable me to conclude whether during that period he continued gambling on course or in betting offices. I consider it probable that one reason for the apparent six week gap in his heavy gambling was that he made an effort to abstain, during a period of relative clarity as to the, by then, exceptional risks which he was running, both for his own and for his family’s security and well being.

Nonetheless, the claimant opened his first betting account with William Hill, numbered 89622V1 by telephone on the afternoon of 2^nd May 2006. I shall refer to it as “Account No 1”. In the space of twenty two minutes either side of 4 o’clock, he had placed four bets of £4,000, £6,000, £20,000 and £18,000 respectively, losing on the first three, but recovering his losses by winning a net £43,200 on the fourth, ending £31,200 ahead of the game. He then stopped.

He repeated the process on the following day, placing four bets during a half hour spanning 9.30 in the evening, with stakes of £1,000, £10,000, £10,000 and £14,000 respectively. Again, he lost on the first three but recovered on the fourth, ending just over £2,000 ahead. In little over a day since opening his account the claimant therefore had staked a total of £83,000 and made net returns slightly in excess of £33,000.

The claimant’s arrival as a new customer of William Hill came to the immediate attention of the Betting Control Department. He had been a successful gambler with Stanley Racing, a bookmaker which William Hill had acquired, and he was therefore identified from the outset as a customer calling for specific monitoring by the Risk Managers. He had also persuaded staff at a Stanley betting office to allow him to bet on credit, to an aggregate of nearly £80,000 on one afternoon in 2005, following which he defaulted in payment, and was excluded by Stanley as a result. This latter episode only came to the attention of Betting Control at William Hill in August 2006, but his general pattern of profitable betting at Stanley was known to Betting Control from the outset. Accordingly they marked his account “Ask all Bets” meaning that telephonists were required to seek specific authorisation for any bet placed by the claimant on the telephone, however small.

Six days later, on 9^th May the claimant indulged, to use his words, in his first telephone betting frenzy with William Hill. Between 4 o’clock and 8.44 p.m. he placed no less than fifteen bets, staking a total of £336,600. Although he remained ahead of the game throughout, he lost on the sixth, seventh, eighth and tenth to fourteenth bets inclusive, recovering his position with a final stake on an evens bet of £79,600, to end the day £139,200 ahead. At that point he said that:

“I had a moment of clarity and horror when I realised what I had done. At that point my account was in credit and I decided to close the account and ask for it never to be reopened again.”

In cross-examination the claimant explained that he had at the time been in London training his greyhound Amarillo Slim for the London Derby where it was (or became) third favourite. This was a major event with trial stages and a final in June, requiring the claimant to be in London for about six weeks in total, to supervise an intensive training and racing programme for the greyhound. He said that his heavy gambling was disrupting his training routine, that he had become disgusted and in a temper with himself, and that he therefore made a snap decision to close his account and request that it be not reopened.

There is a full telephone transcript of his call to the William Hill call centre at 9.35 that evening, just under an hour after the placing of his last bet. He rang the Golden Spurs special number and was transferred by the telephonist to the Golden Spurs Team Leader, Elaine Hudson (“Elaine”). She asked him why he wished to close his account and he replied:

“Just … just too easy to gamble.”

She said:

“Too easy to gamble?”

He said:

“Yeah. Its too … too easy and too convenient and I’ve like won a few quid so …”

Elaine then offered to have Customer Services telephone the claimant to explain William Hill’s Self-Exclusion facility, which she explained as a means of excluding himself from his account for six months, and the claimant declined. The conversation continued as follows:

“EH: So you just want to close the account?

GC: Yeah

EH: Right no problem Mr Calvert. I’ll get that done for you as we speak.

GC: It’s … it’s not to be reopened as well, you know what I mean but …

EH: You don’t want it reopening?

GC: No like. [inaudible] obviously if I could ring up and say I wanna reopen that you know what I mean but …

EH: I’ll put a message on it under no circumstances do you want the account to be reopened.

GC: Yep, yeah. I mean it just … just gets a bit out of hand you know what I mean.

EH: That’s no problem Mr Calvert I’ll put a message on that you don’t want … wish the account to be reactivated.

GC: Yeah.

EH: Alright that’s closed for you now Mr Calvert.”

The claimant then requested and in due course obtained repayment of his net winnings from Account No 1.

Elaine then closed the claimant’s Account No 1, placing on the computerised file the following note in the Additional Information field:

“Closed own req do not re-open clients req”

Elaine did not refer the matter to Customer Services at all. Although that department was closed at the time, there was a procedure for reference by email, as I have described.

As might be expected from her position as leader of the Golden Spurs team Elaine was a senior Team Leader with ten years’ experience as such. Although her note on Account No 1 reflected the substance of the claimant’s request, her response to the claimant’s call involved a number of significant departures from the required procedures. Although she was not called in evidence, Mr Sharkey described a conversation with her manager after the event in which he learned that she had acknowledged that her conversation with the claimant made her aware that he might be a problem gambler. As Mr Sharkey acknowledged in cross-examination, the Social Responsibility Policy and Procedures required all account closure requests relating to problem gambling to be notified to Customer Services, and for that department rather than Team Leaders in the Operations department to make the relevant Vet Code endorsement on the customer’s account, using Vet Code 91 if the self-exclusion procedure had been completed and Vet Code 92 if the account was closed without that procedure being completed. Neither Vet Code notice was placed on the claimant’s account.

Secondly, the procedures required the customer to be told (once referred to Customer Services) that if he did not follow through the self-exclusion procedure and sign a self-exclusion agreement, his account could be reopened by him at any time. Elaine did not explain this to Mr Calvert. On the contrary, the gist of her explanation to him was to suggest that his account would not thereafter be reopened at his request.

The claimant was asked in cross-examination why he had declined the offer of the self-exclusion procedure. He said that he had not heard of that specific procedure before but had successfully excluded himself at Stan James by asking for his account not to be reopened, and had later discovered that this had been successfully implemented by the placing of a message on his account. Furthermore, he said that he had been in a hurry and did not wish to go through a further question and answer process at another call centre (by which he meant Customer Services). He also said that he wanted his account closed permanently, rather than to be subjected to the temptation of reopening it in six months time.

The result of the exchange between the claimant and Elaine was therefore that the claimant went away thinking that he had placed an obstacle in the way of his doing further telephone betting with William Hill should he later succumb to the temptation to do so. It may be that Elaine also thought that her note on his account would provide just such an obstacle. If so, she was mistaken.

The claimant did in fact abstain at least from telephone gambling until 27^th May. He explained this as attributable to having been extremely busy training Amarillo Slim for the London Derby. On 27^th May he decided to try again and telephoned William Hill shortly before 1 o’clock. He asked to reopen his account, and the telephonist Sylvia called it up on her screen for the purpose of checking his debit card details. She was not called to give evidence and although Elaine’s note would by then have appeared on the “Additional Information” field, it is not clear whether Sylvia accessed that field. In any event she reopened his account and passed the claimant to another representative Nicola (whose status was not explained in evidence) to enable him to place a bet on the BMW Golf Tournament, after she had cleared it with Betting Control due to his account being endorsed “Ask All Bets”. The evidence of Mr Eckley from Betting Control suggests that the Additional Information field on the claimant’s account would not have been opened by the relevant Risk Manager on that occasion but that, even if it had been, the Risk Manager would have assumed that the account had been properly reopened for good reason. The claimant staked £30,000 and won a net £36,000 on that single bet.

The claimant placed further bets on Account No 1 on the 28^th and 29^th May, and then on 4^th and 5^th June. Two bets on 28^th May with an aggregate stake of £30,000 yielded a net return of £33,000. Two bets on the 29^th May with an aggregate stake of £80,000 yielded a net return of just under £17,000. He made a single winning bet on 4^th June of £30,000 netting £33,000. In cross-examination he conceded that on none of those occasions was he in a betting frenzy but asserted that he was on the 5^th June. Since the events of that day lie at the heart of the case, it is worth dealing with them in some detail.

In the space of just under an hour, starting at 2.55 p.m. the claimant placed four telephone bets. The first, of £50,000 at 9 to 4 lost. The second was an each way bet of £40,000 which the claimant described as chasing his earlier loss, but it also lost. He then tried betting at longer odds to recover his position. The third, £3,000 at 8 to 1 also lost leaving him in aggregate £93,000 down. He therefore staked £20,000 at 5 to 1, and won, leaving himself £7,000 ahead, having successfully chased his three earlier losses. He was cross-examined about this series of bets, and his reaction to them. The exchange is worth setting out in full:

“Q. On 5th June you had three bets, each of which you lost and then you recouped it at 15:52 having had your first bet just under an hour earlier, and you came out with £7,000 profit and you stopped. Did you get into a betting frenzy on that day?

A. Yes.

Q. You did. When did you get into a betting frenzy on that day, Mr. Calvert?

A. Like I explained earlier, when you are gambler who has a gambling problem, you start chasing losses. That is when you go into this mode and it becomes a frenzy and you start gambling.

Q. So was your first bet of £50,000? Was that in a betting frenzy?

A. No.

Q. Then you had an each way bet about ten minutes later.

A. Yes.

Q. Was that in a betting frenzy?

A. Yes.

Q. Just explain what happened in the ten minutes apart from losing one bet which put you into a betting frenzy?

A. Once you have had a losing bet, you look at the papers straightaway to look for your next bet what you are going to have on.

Q. To do what?

A. Once you have lost £50,000, your reaction is to go and look at the next race to see what you are going to bet.

Q. Yes; but that is deciding that you will have another bet to see if you could get your money back?

A. That is correct.

Q. And you describe that as a ‘betting frenzy’?

A. Yes; because you will keep on going and you will do everything. You will keep on betting to try and get that back, because that is how your mind works. You go into tunnel vision. You don't think, ‘Well, I've just won that the day before or the day before or the week before’. You think, ‘I need to get that money what I have just lost back’. You go into a tunnel and that is when you start chasing in a frenzy and that is when there is a problem.

Q. You see, at this stage, you were still well up on your gambling, were you not?

A. That is right; yes.

Q. The figures show that in the period we have got from October 2005 onwards you were something like a million pounds up at this stage.

A. Yes.

Q. So when you put a second bet on at £40,000, you say that that was a betting frenzy and something you would not have done if you had not been in the frenzy?

A. That is the illness, isn't it? That is the illness. What I am saying is that you can be £10 million in profit and you can have £10 million in the bank, right, but once you lose that £50,000, you will lose that £10 million to try and get that £50,000 back. It is something that is in the mind. When you are like that, that is when you know that you have got a problem. That is how I knew. That was my main downfall and I could not stop chasing. I could not stop, once I had had a losing bet, I had to chase them those losses in my mind, and anything could happen from there. I would just keep on going. When you are like that, you have lost all value for money. You are not thinking about anything other than getting that money back at that stage. When you are thinking that way, you obviously have a problem.

Q. I see. It was following that that you decided to close your account again?

A. Yes.

Q. By which time you were up £125,000 over the period since 27th May and up £298,000 altogether since you had started gambling with William Hill on 2nd May just five weeks earlier?

A. Yes. Like I was explaining to you earlier, this is the problem. This is why I am self-excluding at these points. This is why I am closing the accounts because it would give myself, and on the other accounts, why I did it at that stage is because you go through that, there, right, and at the time your adrenalin is pumping, you have lost £50,000, you have lost £90,000, you are losing, and all of a sudden you are getting of the situation, out of the hole you were in. Then you sit down at the end of the night and you say, ‘God Almighty, what have I done that for? Why am I putting myself through that? Why have I done that?’ But there is no answer to that, other than that you have got a gambling problem. When you are sitting with £300,000 or £400,000 or whatever it may be in that stage, you are not doing it for the money any more. You say to yourself, ‘I'm not doing this for the money. I'm doing this because I am getting a kick out of it or I'm getting a buzz. Why am I doing it?’
This is when I am saying, ‘Right, Graham, you've got a problem. What are you going to do?’ This is when I got on the phone to them and said, ‘Close my account, please. Don't let me re-open it’. On the last occasion, the man said the self-exclusion thing. He was sick of me opening and closing my accounts. The way I understood it was that he was putting a block on it, so no matter what I did I could not... I was putting a block on it myself so I could not come and do this again.”

I accept that thought processes broadly along the lines described in that exchange went through Mr Calvert’s mind before he telephoned William Hill to close his account at 4.38 p.m. that same day. Again, the whole of the relevant conversations were recorded and, (more by luck than by design), the relevant tapes have survived. He began by asking the telephonist Scott how much was in his account. He was told by an obviously surprised Scott that it was £150,000 and he asked for its return. Then he asked for the account to be closed and not reopened. Following the script on which he had been trained, Scott put him through to his Team Leader, John Laird (“John”). It emerged for the first time from disclosure late in the trial of the records of disciplinary proceedings against John that he was at the time only an acting Team Leader. Although he had shadowed another Team Leader, and been monitored by her, he had not been put through the full Team Leader Development Training Programme. He had been involved in the self-exclusion procedure twice before.

Scott put the claimant on hold while he spoke to John, explaining that the claimant wanted his account closed. During that short conversation John accessed the claimant’s account and read the message placed on it by Elaine “Closed on request, do not reopen, client’s request”. He noted that the client had requested reopening on 27^th May, and that the account had been reopened accordingly. He decided to speak to an Operations Manager, but before he could do so the link with the claimant had been cut off. John then had a short inconsequential conversation with an unidentified Operations Manager. Not long thereafter the claimant called again and an unidentified telephonist succeeded in putting him through to John.

Again, it is worth recording the whole of the short but critical conversation which followed, omitting the initial pleasantries:

“JL: Hi Mr Calvert you’re through to John team leader here in Leeds. I understand you want to close the account.

GC: Yes please yeah

JL: Can you tell me why that is please?

GC: Cos it’s just too easy to gamble.

JL: Right. So do you want to self exclude on this point then which means you will not be able to open the account with us again within the next 6 months?

GC: That’s right aye.

JL: Right, well what I’ll do, I’ll pass on all the relevant information.

GC: Right.

JL: The account will now be closed. You will not be able to open it within the next 6 months.

GC: Right and the … the money will be sent back tonight will it? The hundred and … the money what’s in there will be sent back…

JL: I don’t know if it will be sent back tonight or tomorrow. It may be too late for the close of business. We may not get it out tonight. It may be first thing tomorrow morning…

GC: …right…

JL: …but the money will be returned to your account and the account will now be closed for the next 6 months…

GC: …right…

JL: …You will not be allowed to open it under any circumstances. You will not be allowed to bet over the phone with William Hill.

GC: Right okay.

JL: Okay Mr Calvert?

GC: Aye, no that’s fine.

JL: Thank you. Bye bye.

GC: Bye bye. Bye bye.”

A little later John rang the claimant, reminding him of the previous call, and said:

“Yep. Er I just need to double check with you that er it’s okay for our Customer Services to contact you at some point and go over the er script that they’ve got to go over. Is that okay?”

The claimant said that it was, and was left to expect a call the following day.

There is an issue whether at any time the claimant was in fact contacted by Customer Services. The claimant’s case is that shortly after one of the occasions (that is 9^th May or 5^th June) when he had sought to close his account, he was telephoned by Customer Services seeking to persuade him to continue his telephone betting, but upon the basis that he could impose limits on the amount of money which he would be permitted to bet each week. His pleaded case was that this occurred shortly after the 5^th June conversation with John. He made no mention of it in his witness statement, and in cross-examination said that he thought it had occurred after his 9^th May conversation with Elaine. In any event, he said that it had occurred during a time of self-realisation, and that he had declined the invitation.

William Hill’s pleaded response was one of non-admission. No tape recordings of any conversation between the Customer Services department and the claimant have emerged, and neither Elaine nor John passed on the claimant’s requests to close his account and to self-exclude to Customer Services. Mr Starkey established that Elaine had not done so, and it was established in disciplinary proceedings against John in April 2007 that he had not done so either. Furthermore, as will appear, none of the steps or records which, had the matter been passed to Customer Services, ought to have been taken and made in relation to an account closure or self-exclusion were in fact taken or made. I find that no conversation of the type alleged by the claimant took place between him and Customer Services, and further that there was no contact by Customer Services with the claimant at all.

John was not called to give evidence, and until Mr Sharkey revealed for the first time in cross-examination on the third day of the trial that John had been disciplined for his conduct in relation to the claimant, it was wholly unclear what had happened following his call back to the claimant to notify him of an impending call from Customer Services. Lamentably, the detailed records of the disciplinary proceedings against John were not disclosed before the trial, and were made available to the claimant (and in due course to the court) only after Mr Sharkey’s cross-examination, following which he had to be recalled.

That late disclosure revealed that on 31^st March 2007 John was suspended on full pay pending investigation of an allegation that:

“You have failed to follow essential company guidelines in relation to a Social Responsibility incident”

The suspension letter warned him that if the allegation were proved he faced summary dismissal.

The outcome of the investigation was that John was not dismissed, but rather given a final written warning, with consequential loss of 60% of his accrued telephone betting bonus. William Hill’s internal investigation concluded that he had failed to pass on the claimant’s self-exclusion request to Customer Services. It appears likely that he was dealt with relatively leniently because of evidence of his having been under considerable pressure, as appeared from a further tape recording which was also disclosed very late, during the trial.

In my judgment John failed to discharge his responsibilities not merely by failing to pass on the claimant’s self-exclusion request to the appropriate department, but by assuring the claimant, in advance of his having been contacted by Customer Services, and having signed a self-exclusion agreement, that he was, there and then, excluded from telephone betting with William Hill for six months. The claimant was left with the clear impression (entirely contrary to William Hill’s policy and procedures) that he need do nothing further to obtain self-exclusion.

Both John and Elaine also departed from William Hill’s Social Responsibility Policy and Procedures in the following material respects. First, neither of them drew to the claimant’s attention or provided him with William Hill’s responsible gambling leaflet. Secondly neither of them mentioned or encouraged him to contact GamCare. Those requirements were imposed upon staff not merely when a customer clearly indicated that he had a problem with his gambling, or (which is the same thing) requested self-exclusion, but also upon the occasion of any account closure request coupled with an explanation relating to gambling difficulties. As I have recorded, Elaine appreciated that the claimant’s account closure request on 9^th May related to gambling difficulties, and John must plainly have done so as well.

The outcome of the claimant’s self-exclusion request on 5^th June was therefore in fact that his Account No 1 was not closed at all. Elaine’s note remained in its Additional Information field throughout, but the account was otherwise simply left open, after being reopened on 27^th May. Neither Vet Codes 91 or 92 were ever placed on it. In the event however, the claimant made no further use of Account No 1, having by 5^th June staked £702,600 and made net winnings by the use of it £298,391, all in the space of just one month. Of those sums, £283,000 was staked and £125,923 was won after the Claimant reopened his account on 27^th May. He did not however place any further telephone bets with William Hill between 5^th June and 5^th August.

That is not to say that the claimant altogether ceased betting. On the contrary, having travelled to Portugal to inspect a property upon which he placed a deposit (drawn from his winnings), he returned and in late July opened an internet account with Betfair, a company offering a matched bets service. His evidence was that he did so for the purpose of laying bets on a horse to lose. He transferred £90,000 to the account and after laying and having accepted matched bets to an aggregate risk of £40,000, he found that Betfair had intervened to prevent him laying any more. He was told that Betfair had frozen his account while conducting an “unusual betting pattern” investigation due to the size of his bets. Nonetheless the horse duly lost and the claimant made a profit on the exercise in excess of £114,000.

The further history of the Betfair account was the subject of limited and, in my judgment, very unsatisfactory evidence. In his witness statement the claimant said that immediately after the successful bet which I have just described he asked a friend of his to change the password on the account without telling him the new password, so that he would be unable to use it again, it being his understanding that he would not be able to open a new account while the old account remained open. He said that he placed no further bets with Betfair.

In cross-examination he was asked for the name of his friend. His first answer was Craig Mottram, which he then changed to Steven Porter. On further cross-examination after the short adjournment he revealed that Craig Mottram was his brother-in-law, but adhered to his evidence that his friend had been Steven Porter. He said that he was a gambler who had continued to use the Betfair account for betting on dogs, but not on the claimant’s dogs. The claimant displayed obvious reluctance to reveal the name of his friend when asked to do so. It was not put to him that he had used Steven Porter (or for that matter Craig Mottram) as a means of carrying on gambling on greyhounds incognito.

The Betfair episode constituted a very unsatisfactory aspect of the claimant’s evidence, and one about which I have not found it possible to make findings of fact. I shall return to it because of its relevance to the questions of causation and loss. My impression was and remains that there was something about the operation of the Betfair account, and his brother-in-law’s participation in it, which the claimant was most anxious to conceal. In that respect he succeeded, but at serious cost to his credibility, by his obvious lie about the name of his friend.

By late July 2006 the claimant was again firmly in the grip of his urge to gamble (as he put it) and said that he sought to reopen his account at Ladbrokes. He said that he was refused this on the ground of having self-excluded himself. Although his account was Ladbrokes was closed at this time I reject, for the reasons already given, his evidence that this was because he had already self excluded himself from gambling with Ladbrokes. His account had not before July been closed at all. In any event he then opened an account with Coral. On attempting to place a bet of £50,000 he said that he was limited to £500 and, because stakes at that level failed entirely to provide the necessary ‘buzz’, left Coral alone. He then opened an account with Victor Chandler, according to him with the same result, again being limited to £500 and placing no further bets with that bookmaker.

Documents emanating from Victor Chandler suggests that the claimant placed a bet on 10^th August of £10,000 on the golfer Phil Mickleson. In his witness statement the claimant said this must have been a mistake by Victor Chandler. In cross-examination he said that he had tried to put £10,000 on Mickleson with Chandler, but that Chandler had refused, and that he had in fact placed a bet on Mickleson with William Hill on the same day.

In my judgment the combination of evidence from the claimant that he tried to place a bet on Mickleson with Chandler, and documentary evidence from Chandler that such a bet was placed by him on that day makes it probable that the claimant did indeed do so. That conclusion casts grave doubt on his evidence that Chandler limited him to £500, and more generally undermines the whole of his evidence about having his betting severely limited by bookmakers other than William Hill. Again, I shall have to refer to this episode on the question of causation.

There is also evidence from the claimant’s bank statements for the period June to early August 2006 of substantial round sum receipts and payments consistent with continued gambling. The statements in question were first produced by the claimant only when cross examination revealed the existence of the saver account to which they related. The claimant denied that they related to gambling, but could offer no clear explanation for them, other than that they might have related to the purchase and sale of greyhounds. Again, I was not persuaded by this explanation, and it was (of course) wholly unsupported by any documentary records. In my judgment the claimant did more gambling during this period than he has been prepared to reveal, but to an extent that cannot now be quantified.

On 5^th August the claimant resumed telephone betting with William Hill. He did not, as on 27^th May, reopen Account No 1. Rather he sought successfully to open a new account (“Account No 2”). Again, a recording of his conversation with William Hill’s call centre survives. He explained to the telephonist that he had closed his account and wished to reopen the account with a different card, following which the telephonist took the usual details of his name, address, date of birth and bank card number. She then asked whether he could recall his old account number and the claimant replied that he could not. He declined an invitation to receive mail shots and statements through the post but was told that he would receive a welcome pack. After being given his new account number, and a telephone number with which to place bets, he sought unsuccessfully to debit his bank card with £30,000 so he could commence betting. Later that afternoon, in a conversation with another telephonist, he succeeded in making deposits in an aggregate of £25,000 and then sought to place an immediate bet of £12,000 each way on a horse (i.e. a total of £24,000). The telephonist sought and obtained authorisation from the Risk Managers (a transcript of which also survives) and it was provided in circumstances in which (again) the message placed by Elaine on the claimant’s (still open) Account No 1 appears not to have come to the attention of the Risk Manager concerned.

That bet proved to be a loser and later the claimant deposited and bet a further £5,000 at 15 to 2 which also lost, leaving him £29,000 down on the day. This proved to be the first of eight days of telephone betting on Account No 2, concluding on 17^th August. On every one of them the claimant made a net loss, ranging from £10,000 on 17^th August to £124,000 on 8^th August. During that period, every bet placed on each day other than 8^th August lost. Individual stakes ranged between £2,000 and £70,000. The aggregate stakes were £751,000, and the aggregate loss was £416,000.

The 8^th August was the worst day of all, and bears close analysis. The aggregate of five bets (four of which lost) between 3.16 pm and 4.28 pm left the claimant £64,000 down, from which he recovered by a win at 10 to 3 on a stake of £60,000 leaving him £136,000 ahead. Having successfully chased his losses, the claimant did not (as on many previous occasions) stop betting, but continued with a further five losers, including a final desperate stake of an aggregate £70,000 each at 5 to 1 which more than doubled his overall loss until that point. He ended up having staked £459,000 and lost £124,000 on the day.

By this time, the claimant had run out of money. Having enjoyed a healthy cash position in excess of £400,000 in credit on 21^st July, his bank statements show that he was reduced to a cash credit of less than £12,000 by 17^th August, most of which he spent on his 17^th August losing bet of £10,000. He had in the same period also made cash bets in William Hill betting shops amounting in aggregate to £230,000 on which, notwithstanding a single win, he made an overall loss of £153,000.

On 22^nd August the claimant opened a third account with William Hill, (“Account No 3”). This time it was a cash account, which he funded by physical delivery of enormous amounts of cash to William Hill betting shops, often in large sacks, before placing bets with the use of that cash by telephone. He used this account to place bets on eight separate days between 22^nd August and 27^th September. After a net win on 22^nd August, he made losses on every other day, leading to an aggregate loss of £232,000, from aggregate stakes of £634,500. In evidence the claimant said that he raised the cash for his betting on this account by borrowing from friends, while concealing his intended use of the borrowed money. In fact his large scale borrowing started only in September, and the source of his cash deposits prior to starting borrowing was not explained. It may have derived in part from a substantial cash win at a William Hill betting office on 21^st August.

During the same period the claimant continued to place telephone bets on Account No 2, on 15^th, 20^th and 21^st September. For this purpose he also used borrowed money while concealing its intended use. On 15^th September he betted an aggregate £578,900 and made a net loss of £296,900. On 20^th September he betted and lost £120,000 on a single bet. On 21^st September he staked an aggregate £523,333 on seven bets, returning a net loss of £519,999. Three of those bets were cumulative bets on America to win the Ryder Cup, in an aggregate of £347,333. It was the largest bet ever received by William Hill on a golf event, and the company lost no time in publicising both the amount of the bet and the outcome, but without revealing the claimant as the unsuccessful punter.

From 21^st August to the end of September the claimant made a further three visits to place cash bets at betting shops, staking an aggregate £75,000 and, uniquely during this period, making an aggregate net return (as the result of a single successful bet) of £85,000, all of which he said he ploughed back into other loss making gambling.

There is no record of any gambling by the claimant in October 2006. It appears that he temporarily regained a measure of control over his gambling, because on 10^th October he telephoned William Hill to close his accounts, asking for them never to be reopened. He was duly put through to a Team Leader who explained that, if he closed his accounts without going through the self exclusion process with Customer Services, he would be able to reopen them at any time. He chose not to self exclude, preferring simply to close his accounts. Eight days later he reopened them, but placed no further bets that month.

With further borrowed money, the claimant began again betting on his Account No 2 on 1^st, 2^nd and 4^th November, staking an aggregate of £293,700 and returning a net loss of £49,881, with individual stakes in the range £200 (at the end of a day when, no doubt, short of funds) and £50,000. On 17^th November he staked and lost £18,000 on three cash bets at William Hill’s Chester le Street betting office.

Finally, on 1^st and 2^nd December the claimant made the last use of his Account No 2, staking an aggregate of £420,000 and losing a net £180,000 with stakes ranging from £40,000 to £120,000. The claimant recalled losing another £50,000 in cash betting at William Hill’s Fencehouses betting shop in December, albeit that no records of that betting survive.

There is no evidence that the claimant carried on any significant gambling other than with William Hill after opening his Account No 2 on 5^th August other than the isolated bet with Victor Chandler on 10^th August. There is also an uncertainty as to precisely how much cash betting the claimant carried out at William Hill betting offices during the second half of 2006 (or for that matter during any other period), because betting offices do not habitually retain all betting slips, and William Hill had been unable to proffer with any confidence records establishing either the amount or outcome of the claimant’s betting of that type. Such documents as do survive tentatively suggest that during 2006 the claimant made aggregate cash bets at William Hill betting offices of £708,000 odd with a gross return of £875,000 odd, and therefore a net profit of £167,000. The more specific cash bets which I have described above are proved by surviving records, and the overall loss-making picture which they present does not lie happily with the positive aggregate return set forth in an email from William Hill to the claimant on 11^th December 2006, in which the aggregate 2006 figures which I have mentioned above were set out. For reasons which will appear, I have not found it necessary to attempt to resolve that apparent disparity.

A manuscript summary entry for an unspecified period ending on 25^th January 2006 suggests that the claimant placed a total of 331 bets prior to that date at William Hill’s Houghton betting office staking an aggregate £838,257, with receipts of £854,089, and suggesting therefore a net profit of £15,832, and an average individual stake in the region of £2,500. The claimant did not vigorously disagree with the suggestion that this might have been a summary of his betting at that betting office prior to January 2006, but the statistics are of very dubious reliability, as Mr Fenwick accepted.

Having by mid-December 2006 run out both of his own money and of all sources of borrowing from friends, the claimant was forced to discontinue his gambling altogether in mid-December. He claims by then to have borrowed an aggregate of £1,543,500 to fund his gambling during the second half of 2006, which he broke down as £1.42 million borrowed from business associates between August and December, £23,500 borrowed from Lloyds TSB in September and £100,000 raised by a re-mortgage of his home, which he had planned to use for development of the kennels, but in fact used for gambling. Subject to one exception, these amounts were not subjected to specific or successful challenge. The exception relates to two payments amounting in aggregate to £100,000, which passed through the claimant’s current account on the way to a Mr Ivan Danski, and one of which clearly came from one of the claimant’s lenders, a Mr McGinty. I accept Mr Fenwick’s inference that the other payment, making up the aggregate £100,000 probably came from the same source, although that is not demonstrable from the bank statement. Having initially stated that all his borrowings from Mr McGinty were dissipated in gambling, the claimant was constrained to accept that in relation to part of this £100,000 he had in fact used it for payments to Mr Danski, which he described as honouring an earlier commitment to make loans to him. These formed part of a much larger series of loans by the claimant to Mr Danski, who he described as a garage proprietor in Newcastle, in an overall aggregate of £497,000, all of which save for £100,000 odd, he said that Mr Danski repaid during 2006 and 2007.

The claimant’s evidence, the substance of which I accept, was that he concealed at least the extent and adverse consequences of his gambling both in the second half of 2006 and earlier from his wife, his mother and other members of his family until December 2006, when he confessed what he had done. With hindsight, he said that he suspected that his family began to suspect the truth, particularly at the time when publicity was given to his Ryder Cup bet, albeit with no express reference to him personally. His wife has since left him, and the claimant attributes this to his excessive gambling. There is however evidence from the claimant himself that their relationship had begun to break down in mid-2006, for reasons entirely unconnected with his gambling.

The claimant also gave up his greyhound trainer’s licence in December 2006, and it was taken over by his mother, who continued to run the greyhound training business on a much reduced scale through 2007, with what the claimant described as modest administrative assistance from him in the background. It emerged in cross-examination that the claimant had received a fine and a severe reprimand from the greyhound racing regulatory body in October 2006, in connection with his failure to check whether a skin allergy cream and anti-histamine tablets administered to a greyhound contained traces of a prohibited performance-enhancing painkiller, and that a further inquiry in relation to his greyhound Amarillo Slim was opened late in 2006, which remains unresolved. Nonetheless, the claimant denied that these events had anything to do with his giving up of his greyhound trainer’s licence, which he attributed to his financial ruin coupled with publicity as to his excessive and improvident gambling, which he described as fatal to his continued reputation as a reliable trainer. On balance I accept that explanation.

Both the government White Paper and the 2005 Gambling Act which followed it identified the class of persons calling for special attention due to their vulnerability to the temptation to gamble as ‘problem gamblers’. They did so either by using that phrase as a term of art or by the broader phrase in section 24, requiring assistance to be made available to “persons who are or may be affected by problems relating to gambling”.

The claimant’s case that he was owed a duty of care by William Hill is based in part upon the allegation that he was a problem gambler by the time he commenced betting with William Hill. Paragraph 3 of the Particulars of Claim sought to define problem gambling as “a disorder in which the addict feels an irresistible urge to place bets”. William Hill put the claimant to proof both of the existence of such a gambling disorder, and of his claim to have suffered from it at any material time.

Both parties called experienced psychiatrists as expert witnesses. Both produced helpful reports, and after a meeting followed them up with a joint report dated 14^th February 2008. The claimant called Dr Kayetan Zakrzewski MRCPsych and the defendant called Dr John Frazer MRCPsych. I found them both to be wholly independent witnesses, fully cognisant of their duties to the court. They were both determined to, and did, provide considerable assistance.

Dr Zakrzewski is a consultant psychiatrist with thirty years’ experience in psychiatry, for twenty years as a consultant, in Poland and in England. During the last six years he has tended to specialise in addictive disorders. Within that specialisation problem gamblers have constituted a small proportion of his overall work, amounting, to his best recollection, to about twenty to thirty in all. He said, and I accept, that his specialising in addictive disorders meant that he regarded it as likely that he had a greater professional exposure to problem gamblers than would be typical of consultant psychiatrists generally. As an expert witness Dr Zakrzewski’s value as an experienced specialist was slightly undermined by a lack of focus in his response to questions in cross-examination. Furthermore, as will appear when I address the few issues which separated his opinion from that of Dr Frazer, I did not find that Dr Zakrzewski’s opinion as to the duration and gravity of the claimant’s condition was fully supported by the process of logical analysis with which he sought to support it.

Dr Frazer began his professional career as a GP, between 1985 and 1997. Since then he has specialised entirely in forensic psychiatry, and his experience of gambling addicts (at the rate of about two a year, amounting to approximately twenty five in all) has arisen mainly in circumstances where such persons have been referred to him for report or treatment where they have committed, or been accused of having committed, criminal offences. Dr Frazer was a highly articulate witness, who was prepared to and did modify his opinion in the light of his developing understanding of the underlying facts, and in response to cross-examination. His classification of the gravity of the claimant’s condition was inevitably coloured by his particular experience of assessing and treating patients in the context of actual or suspected crime. Thus for example his view that the claimant was not particularly seriously affected by a gambling disorder was attributable in no small measure to the fact that, by contrast no doubt with some of his patients, the claimant had not resorted to theft or robbery in order to feed his gambling habit.

To a very large extent the two psychiatric experts concurred in their opinions both as to the appropriate classification and relevant tests for the diagnosis of gambling disorder. To the extent that they differed, I did not find the evidence of one to be uniformly preferable to that of the other.

Before expressing my conclusions on the psychiatric issues, it is necessary for me to say a little more about how they developed during the course of the preparation and trial of the case. Dr Zakrzewski’s written report, based upon a lengthy interview with the claimant in October 2007, a study of his medical records and of the pleadings, concluded that he had been a pathological gambler, within the meaning of that phrase adopted by the American Psychiatric Association, to a severe extent throughout the relevant period, by which he meant 2006.

Dr Frazer based his report upon an examination of the claimant on 4^th January 2008, together with a substantial reading of the papers generated by the litigation. His opinion was that, while the claimant was a problem gambler, which he described as a form of Impulse Control Disorder, his symptoms fell short of those which would classify him as a pathological gambler (adopting the same American Psychiatric Association definition).

In their joint report dated 14^th February 2008 Dr Zakrzewski adhered to his previous opinion, while Dr Frazer acknowledged that, at certain times rather than continuously, the claimant’s symptoms corresponded with those of a pathological gambler. His opinion was that the claimant’s gambling was “out of financial control” only for a short period in 2005, and then between May and August 2006.

When this difference was examined in more detail in cross-examination, in particular by reference to the evidence as to the claimant’s patterns of gambling during 2006, Dr Zakrzewski maintained his original opinion, while Dr Frazer eventually accepted that during the period from June to November or December 2006, the claimant exhibited most of the relevant criteria for the identification of a pathological rather than merely problem gambler.

This state of the evidence gave rise to an application by the claimant to amend his Particulars of Claim, the effect of which was to include for the first time in a claim which had until then sought compensation for pure economic loss, a claim for damages for personal injury constituted by the deterioration in his gambling disorder caused by William Hill’s alleged breach of duty, in particular in permitting him to continue to gamble, after his attempt at self-exclusion. The amendment pleaded in the alternative that this had caused his deterioration from problem to pathological gambler, or alternatively an aggravation of an existing pathological gambling disorder, in the second half of 2006. Furthermore, (as correctly anticipated by Mr Fenwick when opposing the claimant’s application for permission to amend) the amendment opened the way for the claimant to argue that, on the basis that the deterioration in his gambling disorder was a contributory cause of his gambling losses, that those losses were not pure economic loss, but financial loss consequential upon the suffering of personal injury (in this case psychiatric injury). I allowed the amendment, for reasons given in a short judgment at the time.

By the conclusion of their cross-examination, there was a large measure of common ground between Doctors Zakrzewski and Frazer, which may be summarised as follows. First, both problem gambling and pathological gambling are aspects of a psychological disorder, the central feature of which is that, without necessarily losing legal capacity to gamble, the patient’s ability to make reasoned and informed judgments about his gambling is impaired. The impairment arises from a behavioural rather than physical addiction, best summarised as an emotional urge to gamble. The condition is therefore described as an Impulse Control Disorder.

Secondly, the condition is properly described as a spectrum disorder. At the less serious end of the spectrum (problem gambling) the patient experiences increasing difficulty in controlling his gambling. At the more serious (pathological) end of the spectrum, the disorder makes control of his gambling impossible. This is reflected in the following passage in the joint report:

“Dr Zakrzewski agreed with Dr Frazer that there was a distinction between people who were able to control their gambling, although finding it hard to do so, and those unable to control their gambling in the sense that it was a Spectrum Disorder with low severity at one end and high severity at the other end and that “problem gambling” would fit towards the low end of severity and “pathological gambling” towards the extreme end.”

Thirdly, the disorder is properly characterised as both continuous and repetitive. This is because the patient generally experiences bouts of severe impairment of control, separated by moments or periods of relative clarity during which he may temporarily recover control, while remaining vulnerable all the time to relapse.

Fourthly, the most reliable and internationally accepted criteria for the diagnosis for pathological gambling are those to be found in the DSM-IV tests developed over many years by the American Psychiatric Association, and approved by the BMA Board of Science. I shall return to those in due course, but in summary they consist of ten diagnostic criteria. When three or more of them are met, the patient is diagnosed as a problem gambler. When five or more are met, the patient is diagnosed as a ‘probable pathological gambler’.

Fifthly, at no time did the claimant suffer a loss of legal capacity to gamble, or suffer from psychotic delusions. Sixthly, he was throughout 2006 at least a problem gambler. Finally from June to December 2006 he met sufficient of the DSM-IV diagnostic criteria to be properly classified as a probable psychological gambler. Nonetheless Dr Frazer was reluctant in cross examination to concede that the claimant had altogether lost control of his gambling for the whole of the period between June and December 2006 inclusive.

It might be thought from that summary that the main issue as to the claimant’s psychological condition which I have to decide was simply whether he had deteriorated from being a problem gambler to a pathological gambler earlier in 2006 than in June. Undaunted by the apparent agreement of the experts that the claimant was a probable pathological gambler from June onwards, Mr Fenwick submitted that I should nonetheless conclude, upon the whole of the evidence, that the claimant retained sufficient control of his gambling throughout the events which I have described that he never became a pathological gambler at all, in the sense of his gambling being out of his control. The true view was, he submitted, that by mid-2006 the claimant had been for many years a highly successful and professional gambler, motivated by a desire to make profits to support his personal and business life, rather than by any emotional urge, whose only misfortune thereafter was that his luck ran out. He submitted that the apparent consensus of the psychiatric experts to the contrary arose from them both having accepted uncritically the claimant’s account of his misfortunes while being examined by them, and that the rigorous forensic examination of the claimant’s evidence made possible by a full trial both could and should enable the court to reach a contrary conclusion. He reminded me in passing that in Foroughi v. Star City Pty Ltd [2007] FCA 1503 Jacobson J (sitting in the New South Wales District Registry of the Federal Court of Australia) had accepted expert evidence that a gambler who satisfied five of the DSM-IV criteria might nonetheless not in fact be a pathological gambler, and he relied on the undoubted fact that the DSM-IV criteria are expressed in sufficiently plain factual language to permit them to be applied to the facts as found by a judge without the need for, still less the obligation to be bound by, expert psychiatric assistance.

The DSM-IV criteria (as they appear in a publication by the BMA Board of Science entitled ‘Gambling addiction and its treatment within the NHS’ issued in January 2007) are as follows:

“DSM-IV Diagnostic criteria for pathological gambling

The updated DSM-IV consists of 10 diagnostic criteria. A ‘problem gambler’ is diagnosed when three or more of criteria A1-A10 are met, and a score of five or more indicates a ‘probably pathological gambler’. The diagnosis is not made if the gambling behaviour is better accounted for by a manic episode.

A.

Persistent and recurrent maladaptive gambling behaviour as indicated by five (or more) of the following:

(1)

is preoccupied with gambling (eg preoccupied with reliving past gambling experiences, handicapping or planning next venture, or thinking of ways to get money with which to gamble)

(2)

needs to gamble with increasing amounts of money in order to achieve the desired excitement

(3)

has repeated unsuccessful efforts to control, cut back, or stop gambling

(4)

is restless or irritable when trying to cut down or stop gambling

(5)

gambles as a way of escaping from problems or of relieving a dysphoric mood (eg feelings of helplessness, guilt, anxiety, depression)

(6)

after losing money gambling, often returns another day to get even (‘chasing’ one’s losses)

(7)

lies to family members, therapist, or others to conceal extent of involvement with gambling

My findings as to the claimant’s psycological condition are as follows. For the whole of the period 2000 to 2004 the claimant was a skilled professional gambler, motivated by a combination of the ability to make substantial profits to support his business and personal life and the ‘buzz’ he obtained from gambling, and who sensibly focused his gambling upon events with which he was most experienced, namely greyhound racing.

During 2005 two developments in his gambling exposed him to the grave risk of becoming a problem gambler. The first was a transition from betting predominately on greyhounds to betting on horses and other sporting events such as golf and football in which he was much less experienced, a move probably provoked by the tendency of bookmakers to limit or prohibit his betting with them on greyhounds, as his stature as a young greyhound trainer increased. Whether he diversified rather than reduce his gambling to maintain his profits or to satisfy his urge I do not know, but it was probably a combination of both.

The second change consisted of his commencement of remote, mainly telephone, gambling. This greatly increased the risk of a loss of control because, as the claimant himself put it, handing over cash at a betting office helps maintain some sense of the financial consequences of gambling, whereas betting numbers on the telephone does not. Furthermore, remote gambling naturally releases the gambler from the practical constraints of obtaining large quantities of cash before visiting a betting office, as the claimant vividly described.

The claimant became a problem gambler by the beginning of 2006. There was a marked increase in the size of his stakes during the latter part of 2005 which he attributed (and I believe him) to a need to bet larger amounts to satisfy his desire to obtain what he called a ‘buzz’. This satisfies DSM-IV test (2). I am also satisfied that by the end of 2005 he had become habituated to concealing the scale of his gambling from his family, albeit that this is corroborated in documents only in May 2006, when he asked William Hill not to send any correspondence relating to his gambling to his home address. This satisfies DSM-IV test (7).

There is in my judgment sufficient evidence of ‘chasing losses’ in his telephone betting with Pagebet and Stan James to satisfy DSM-IV test (6) in substance. Sometimes he successfully chased losses within the day. On other occasions he did so on later days. Generally speaking, he did so with considerable success, but in my judgment continued success is by no means fatal to a conclusion of substantially impaired control, or even loss of control. A skilful controlled gambler may have a run of losses, just as a problem or pathological gambler may have a run of wins.

In my judgment the claimant had become a pathological gambler by May 2006. In addition to continuing to satisfy the DSM-IV criteria (2), (6) and (7), I consider that he had also by the end of May satisfied criteria (3) and (9). As to (3), I have found that he had made attempts to close his account with Stan James so that it should never be reopened. He clearly did so in his telephone conversation with Elaine at William Hill on 9^th May. As to criterion (9), the sheer scale of his betting (both in terms of frequency and amount) by comparison with the value of his property and the risks of financial ruin thereby incurred plainly put his relationship with his family in jeopardy (since he had concealed the scale of his gambling from them). Furthermore, his evidence (which I accept) was that a reason for his seeking permanently to close his William Hill account on 9^th May was the adverse effect which his preoccupation with gambling was having upon his attention to the highly important task of training his greyhound Amarillo Slim for the London Derby.

The claimant therefore scored five on the DSM-IV diagnostic test by May 2006, sufficient to justify his classification as a ‘probable pathological gambler’. In one sense, that probability is a sufficient test for the court. Nonetheless I acknowledge that there may be cases in which an apparent score of five on the DSM-IV tests may be attributable in whole or in part to special factors which, upon close analysis, lead to a contrary conclusion. Having considered Mr Fenwick’s submissions, I am not persuaded that any of them justify such a contrary conclusion. In my judgment the minute analysis of the claimant’s conduct during the course of an eight day trial confirms rather than detracts from a view that the claimant was by the end of May a pathological gambler. I am fortified in reaching this conclusion by Dr Frazer’s substantially identical conclusion. The end of May lies in substance half way between the view which he expressed in the joint report (May onwards) and the view at which he arrived finally in cross-examination (June onwards).

A conclusion that a patient’s condition has deteriorated to the state of a pathological gambler, by reference to his manifestation of the requisite symptoms by a certain date by no means compels a conclusion that he did not in fact reach that state earlier. In that context I bear in mind Dr Zakrzewski’s opinion maintained from start to finish, that the claimant had become a serious pathological gambler considerably earlier in 2006. The heart of Dr Zakrzewski’s analysis may be found in paragraphs 12.2 to 12.3 of his report. After listing an alternative set of diagnostic criteria (based on the Oxford Textbook of Psychiatry, 2000 edition) he said that:

“In my view Mr Calvert displays all the above characteristics to a severe degree and consistently throughout the material time.”

Dr Zakrzewski was unable to justify that assertion in cross-examination and his preference, in agreement with Dr Frazer, for the DSM-IV criteria as the best diagnostic guide detracts from the force of the analysis in paragraphs in 12.2 and 12.3 of his report in any event. For those reasons I am confirmed in the conclusion which I have reached, notwithstanding Dr Zakrzewski’s opinion to the contrary. As will appear, it is not critical to my analysis of the legal issues whether the claimant’s condition deteriorated to that of a pathological gambler in, or shortly before May 2006.

It is in my judgment clear that the claimant’s gambling condition continued to deteriorate after May 2006, in particular once he ran out of money of his own in August. While continuing to satisfy the criteria to which I have already referred, I consider that he also satisfied criterion (1) (thinking of ways to get money with which to gamble) and (10) (relying on others to provide money to relieve a desperate financial situation caused by gambling) from August onwards. Furthermore, he satisfied several of the other criteria to a more substantial extent as time went on, and in particular (2), (6), (7) and (9).

For these reasons I consider that Dr Zakrzewski was correct to describe the claimant as a severe pathological gambler who had lost control of his gambling (rather than merely suffered an impairment of his control) during the last quarter of 2006. In this respect, while I acknowledge Dr Frazer’s point that, by contrast with cases within his experience, the claimant did not at any time resort to crime in order to feed his addiction, I consider that he placed too much emphasis on that factor as a reason for his conclusion that the claimant never became a severe pathological gambler, and never fully lost control of his gambling. I can well understand why, from his particular experience in forensic psychiatry, he may genuinely have come to this conclusion, but in that respect I consider that Dr Zakrzewski’s more general experience, and his specialism in addictive disorders, provides the more reliable guide, and fortifies a conclusion reached by reference to a common sense application of the DSM-IV criteria. In my judgment it would be flying in the face of reality and common sense to reach any other conclusion than that the claimant had lost control of his gambling by the last quarter of 2006. The evidence of a person carrying sack loads of borrowed cash into betting shops in order to fund telephone gambling at ever increasing stake levels, and continuing to do so in the face of unremitting losses to the complete ruin of his career and the security of his family, admits of no other sensible conclusion.

In that respect I reject Mr Fenwick’s submission that the combination of continued gaps between bouts of gambling and the evidence that the claimant made a substantial loan to a business colleague during the relevant period points significantly away from that conclusion. Even severe pathological gambling is a condition which admits of moments of clarity and rationality, and of other uses of available money.

The English courts have never addressed the question whether, and if so in what circumstances, a bookmaker may owe a common law duty of care to a customer known or suspected to be a problem gambler. As originally pleaded, the present case alleged only pure economic loss, so that the duty of care question fell squarely within the category for which the House of Lords recently reviewed the applicable principles in Customs and Excise Commissioners v. Barclays Bank Plc [2006] UKHL 28; [2007] 1AC 181. The late amendment to plead that William Hill’s conduct caused or aggravated the claimant’s condition as a pathological gambler requires the duty of care issue to be addressed not merely in relation to the pure economic loss constituted mainly by gambling losses, but also in relation to personal injury. Since Page v. Smith [1996] AC 155, it is no answer to a claim for damages for personal injuries caused by negligence that the injury in question is purely psychiatric, even though there is still substantial doubt whether foreseeability of purely psychiatric (rather than physical) injury in a primary victim is sufficient to give rise to a duty of care: see White v Chief Constable of South Yorkshire [1999] 2 AC 455, (where it was not), and McLoughlin v Jones [2002] QB 1312, (where, in a contractual relationship, it was).

Understandably, in the light of the modest amount claimed in respect of psychiatric injury by comparison with the large claim in relation to economic loss, the parties spent little time on the legal analysis of the former, concentrating their focus on the latter, with copious reference to authority. I intend to take the same course. I am encouraged in doing so by the observation of Brooke LJ in McLoughlin v Jones, at paragraph 28, that in cases of pure psychiatric injury the principles applicable to pure economic loss may be of real value in any event.

It is now well established, since at least Marc Rich & Co AG v. Bishop Rock Marine Ltd [1996] 1 AC 211 that, in relation to a novel category such as the present, the imposition of liability for negligence must satisfy the three stage test of foreseeability, proximity and fairness. In relation to liability for pure economic loss, the principal conclusion of the House of Lords in the Barclays Bank case was that no single common denominator could be identified among the available tests used in considering whether a defendant owed a duty of care, but rather that each case depended upon its detailed facts, and the particular relationship between the parties in the context of their legal and factual situation taken as a whole. Nonetheless, in five separate but mainly concurrent speeches their Lordships gave detailed analysis of the three main types of test, and their inter-relationship in particular cases, the three types being: (1) voluntary assumption of responsibility; (2) the three stage test already described; and (3) the incremental approach.

For present purposes the most valuable insights provided by the speeches in the Barclays Bank case relate to the true meaning and nature of the assumption of responsibility test, and to the types of case in which it may be a sufficient, albeit not necessary, test for the establishment of a duty of care. In my judgment the Barclays Bank case establishes the following principles in relation to the application of the voluntary assumption of responsibility test, material to the analysis of the present case.

First, the assumption of responsibility test is at its most influential, and may on its own be decisive, where the relationship between the parties has “all the indicia of contract save consideration” (per Lord Bingham at paragraph 4).

Secondly, although the paradigm case of the making of a statement or the giving of advice to a person likely to rely upon it is one where a voluntary assumption of responsibility will be all that is needed to establish a duty of care, there are other fact situations where the test may still be useful, even though no reliance upon information given can be shown. As Lord Hoffmann put it at paragraph 38:

“Even in this context, however, the notion of assumption of responsibility serves a different, weaker, but nevertheless useful purpose in drawing attention to the fact that duty of care is ordinarily generated by something which the defendant has decided to do: giving a reference, supplying a report, managing a syndicate, making ginger beer.”

Thirdly, the question whether a defendant has voluntarily assumed responsibility is to be ascertained objectively, that is, by a consideration of what he did or said rather what he thought or intended: see per Lord Bingham at paragraph 5 and per Lord Hoffmann at paragraph 35.

Fourthly, the word “voluntary” means conscious, considered or deliberate rather than without compulsion or remuneration: see per Lord Walker at paragraph 73. A decision to do something is no less voluntary because it has been done pursuant to a perceived statutory duty or code of conduct: see again per Lord Hoffmann at paragraph 38 where, continuing from the passage which I have cited above, he said:

“It does not much matter why he decided to do it; it may be that he thought it would be profitable or it may be that he was providing a service pursuant to some statutory duty, as in Phelps v. Hillingdon London Borough Council [2001] 2 AC 619 and Ministry of Housing and Local Government v. Sharp [1970] 2 QB 223.”

Finally, it is not a necessary element in a finding that there has been an assumption of responsibility that the defendant should have assumed legal responsibility, that is liability for damages in the event of performing the task carelessly. This appears from Lord Mance’s speech at paragraph 92, referring to Lord Browne-Wilkinson’s speech in White v. Jones [1995] 2 AC 207:

“Lord Browne-Wilkinson, at pp 273G-274G addressed the doubts expressed by Lord Griffiths in Smith v. Eric S Bush and Lord Roskill in Caparo Industries Plc v. Dickman by explaining assumption of responsibility as ‘assumption of responsibility for the task not the assumption of legal responsibility.’ He said:

‘If the responsibility for the task is assumed by the defendant he thereby creates a special relationship between himself and the plaintiff in relation to which the law (not the defendant) attaches a duty to carry out carefully the tasks so assumed.’”

But as Lord Mance continued, at paragraph 93:

“But if all that is meant by voluntary assumption of responsibility is the voluntary responsibility for a task, rather than of liability towards the defendant, then questions of foreseeability, proximity and fairness, reasonableness and justice may become very relevant…. Incrementalism operates as an important cross-check on any other approach.”

Since in the present case William Hill at no time expressly assumed legal responsibility to the claimant, this last dictum of Lord Mance is in my judgment of particular significance in the present case.

The Barclays Bank case concerned loss suffered by the claimant due to the careless failure by the Bank to prevent its customer drawing money from an account after being notified of the making of a freezing injunction prohibiting such conduct by the customer. Although it might be said to fall within the category of cases where a duty was alleged (unsuccessfully) to prevent the claimant being harmed by a third party, it was far removed from the present case, where the prime cause of the claimant’s loss, whether psychiatric or financial, was his own decision during the second half of 2006 to continue his gambling; i.e. self-inflicted loss. In such a context, as the citation from Reeves v. Commissioner of Police at the beginning of this judgment makes clear, the starting point is that the law imposes no general duty upon a person to prevent his neighbour from harming himself. As Lord Hoffmann put it:

“It reflects the individualist philosophy of the common law. People of full age and sound understanding must look after themselves and take responsibility for their actions.”

See also per Lord Hoffmann again in Tomlinson v. Congleton Borough Council [2004] 1 AC 46 at paragraphs 44 to 46, under the heading ‘free will’.

In all the cases where such a duty has been identified, there have been found to be special circumstances which justify a departure from that general rule. In Reeves itself, the (admitted) duty was owed by a police authority to a prisoner who committed suicide when in its care by reason of the complete control which prison authorities have over their prisoners, combined with the special danger of people in prison taking their own lives.

In Jebson v. Ministry of Defence [2000] 1 WLR 2055 the Court of Appeal held that a duty of care was assumed in the connection with the provision of transport by the Army of soldiers to and from an ‘evening out’ on return from which a drunken soldier suffered serious injury when falling out of the back of an open lorry while trying to climb onto the roof. The defence was that the soldier had only himself to blame for getting drunk. At paragraph 25, Potter LJ said:

“[Counsel for the MOD] has also rightly emphasised that, in the ordinary way and in most situations, an adult (and these young men were adults) is not entitled to pray in aid his own drunkenness as giving rise to a duty or responsibility in others to exercise special care. However, that is not an invariable rule: nor is it one which it is fair just and reasonable to apply in circumstances where an obligation of care is assumed or impliedly undertaken in respect of a person who it is appreciated is likely to be drunk.”

In Watson v. British Boxing Board of Control Ltd [2001] 2 WLR 1256 the claimant was seriously injured in a professional boxing match governed by rules established by the defendant, pursuant to which ringside medical facilities were made available, but which fell short of providing immediate resuscitation. By the time he received resuscitation in hospital he had sustained permanent brain damage which ringside treatment would have prevented. The boxer’s decision to fight was of course the primary cause of his brain damage. Although the Board was not the promoter of the fight, it had what Lord Phillips MR described at paragraph 87 as:

“Not merely a measure of control, but complete control over and a responsibility for a situation which would be liable to result in injury to Mr Watson if reasonable care was not exercised by the Board.”

The Board was therefore held liable for carelessness in its formulation of the regulatory regime for ringside medical assistance.

In so concluding, the Court of Appeal relied on its own earlier decision in Barratt v. Ministry of Defence [1995] 1 WLR 1217, in which a naval rating had rendered himself insensible by excessive drinking, and who then died from inhaling his own vomit while being inadequately cared for under the supervision of his petty officer. Beldam LJ held, at page 1225:

“Until he collapsed, I would hold that the deceased was in law alone responsible for his condition. Thereafter, when the defendant assumed responsibility for him, it accepts that the measures taken fell short of the standard reasonably to be expected. It did not summon medical assistance and its supervision of him was inadequate.”

These and other cases show that exceptional circumstances may give rise to a common law duty of care to prevent or to mitigate the consequences or aggravation of self-inflicted harm. Such circumstances may include the assumption of control over a person while vulnerable to the consequences of self-inflicted harm, or the assumption of some responsibility for the care of, or the provision of assistance to, such a person. In every such case the three stage test will be an important part of the analysis whether the circumstances are sufficiently exceptional.

There are no less than four reported cases in which the Australian courts have addressed the question whether the provider of gambling facilities owes a duty of care to a problem gambler. All four of them had recourse, directly or indirectly, to a restatement of general principles regulating the identification of a duty of care to prevent economic loss laid down by the High Court of Australia in Perre v. Apand Pty Ltd [1999] CLR 180, a decision as important in the development of Australian law as the Barclays Bank case has been in England. In order fully to understand the Australian gambling cases, it is necessary first for me to say something about Perre v. Apand itself. The defendant potato crisp manufacturer had through carelessness introduced a form of potato disease onto the land of one of its growers in South Australia. Although the disease was not shown to have spread, neighbouring farm owners suffered economic loss by the imposition of a potato marketing ban in Western Australia attributable to the proximity of their farms to the outbreak of the disease, and sued the defendant for what was therefore pure economic loss (the absence of any escape of the disease preventing a claim under Rylands v. Fletcher).

For present purposes, two themes in the judgments of the members of the High Court are relevant. The first, propounded by McHugh J, was that an important criterion for the imposition of liability for economic loss lay in ascertaining the extent to which the plaintiff was vulnerable to incurring loss by reason of the defendant’s conduct, and the extent to which that was or should have been apparent to the defendant. At paragraph 104 he said this:

“What is likely to be decisive, and always of relevance, in determining whether a duty of care is owed is the answer to the question, “how vulnerable was the plaintiff to incurring loss by reason of the defendant’s conduct?” So also is the actual knowledge of the defendant concerning that risk and its magnitude. If no question of indeterminate liability is present and the defendant, having no legitimate interest to pursue, is aware that his or her conduct will cause economic loss to persons that are not easily able to protect themselves against that loss, it seems to accord with current community standards in most, if not all, cases to require the defendant to have the interest of those persons in mind before he or she embarks on that conduct.”

Later, at paragraph 118, he described vulnerability of that type as “ordinarily a prerequisite to imposing a duty”. In that case the vulnerability of the plaintiffs was total.

The second theme, propounded in particular by Gauldron J, looks first at the question whether the act complained of gives rise to the plaintiff’s loss or impairment of a legal right. At paragraph 38 she said:

“Where a person is in a position to control the exercise or enjoyment by another of a legal right, that position of control, and by corollary, the other’s dependence on the person with control are, in my view, special factors or, which is the same thing, give rise to a special relationship of “proximity” or “neighbourhood” such that the law would impose liability on the person with control if his or her negligent act or omission results in the loss or impairment of that right and is, thereby, productive of economic loss.”

Taking the Australian gambling cases in chronological order, the first, Preston v. Star City Pty Ltd [1999] NSWSC 1273 was a case in which the defendant casino failed to strike out a claim on the basis that there could have been no duty of care owed. It is therefore of limited authority. The defendant casino operator was alleged to have known that the plaintiff was a problem gambler. Wood CJ held it to be at least arguable that the defendant owed the plaintiff a duty not to encourage him to gamble beyond the limit which he had asked the casino to set.

Of much more importance is Reynolds v. Katoomba RSL All Services Club Ltd [2001] NSWCA 234, a decision of the New South Wales Court of Appeal. The plaintiff was a member of the defendant gambling club and the trial judge found that the defendant knew or ought to have known that the plaintiff was a problem gambler. The duty of care alleged was to advise the plaintiff to resign his membership, to warn him of his lack of prudence in gambling and to refuse to permit him to cash cheques or to make cash advances to him for gambling purposes. The trial judge dismissed the claim.

In dismissing the appeal, the main ground relied upon was that, on the evidence, the plaintiff’s gambling problem was insufficient to prevent him from retaining both full responsibility and sufficient control over his urge to gamble: see in particular per Powell JA at paragraph 127 and per Giles JA at paragraphs 147 to 150, culminating in this rhetorical question:

“Even with knowledge of problem gambling, how is the club to know, when asked to cash a cheque, whether the anticipated gambling is the unwanted but compulsive craving of a problem gambler or the choice of a sometime problem gambler then in control of his indulgence?”

Agreeing with Powell JA, Spigelman CJ went further. In his view, gambling losses should not be recognised as a form of loss for which the law permits recovery, save in an extraordinary case the like of which he could not envisage: see paragraph 9. Protection from gambling losses was not a right or interest sufficient to attract protection of the law: see paragraphs 15 to 16. In particular, he regarded as falling outside the protection of the law any economic loss arising from the plaintiff’s own voluntary act.

Thirdly, Spigelman CJ regarded the identification of a duty of care to protect problem gamblers from gambling losses as an infringement of the gambler’s autonomy, a principle he found reflected in Reeves v. Commissioner of Police, as well as in Perre v Apand. After citing the passage in Lord Hoffmann’s speech with which I began this judgment, he continued:

“However, the interest being protected in that case was the risk of physical injury. Circumstances in which there is a duty to protect from the self-infliction of economic loss must, at the very least, be even rarer. Indeed as presently advised, I am unable to conceive of such a case.”

Reynolds v. Katoomba was applied in Forourghi v. Star City Pty Ltd [2007] FCA 1503, to a case in which a problem gambler had signed a self-exclusion agreement, which contained an express release of the defendant against all liabilities. He continued gambling and failed to seek counselling and advice, in both cases in breach of undertakings to do so contained in the exclusion agreement. He then sought to recover subsequent losses on the basis of the casino’s failure to exclude him. Jacobson J found that there had been no duty of care, no breach, and that he was the cause of his own misfortunes.

This was a case in which the plaintiff had, according to expert evidence, scored five on the DSM-IV tests, but in which the judge nonetheless accepted the defendant’s expert’s evidence that, on the facts, the plaintiff was at no time a pathological gambler and that, even if he had been, “such persons can exercise control and limit or cease gambling if they choose to do so”. In my judgment, although the nearest on its facts to the present case, it adds little to the reasoning in Reynolds v. Katoomba and the court’s conclusions as to the extent to which the plaintiff retained the necessary control appear to me to have turned upon the evidence in that case, rather than upon the recognition of any general legal principle.

The final Australian decision, Harry Kakavas v. Crown Ltd [2007] VSC 526 adds nothing further, save for the observation by Harper J, at paragraph 46, that a casino operator might commit a breach a duty of care where, knowing that the customer was a problem gambler, it carelessly failed to have that knowledge in mind, when taking steps that amounted to unwitting exploitation of the customer.

Undeterred by the generally discouraging thrust of the Australian gambling cases, Miss Day puts the claimant’s case for the existence of a relevant duty of care in two ways, one broad and one much narrower. Taking the broad route first, she submitted that the common law should now recognise a duty of care on the part of a bookmaker to any customer appearing to be a problem gambler, first to take reasonable steps to offer assistance, including but not limited to a self-exclusion arrangement, and second to avoid unconsciously (by carelessness) exploiting the customer’s vulnerability by permitting that customer to continue gambling.

Miss Day submitted that this duty of care arose fairly and squarely from the recognition as a matter of public policy that the small defined class of problem gamblers needed special protection, as reflected in the March 2002 White Paper to which I have already referred, the licensing objectives in section 1 of the Gambling Act 2005, the November 2006 Code of Practice issued by the Gambling Commission, and the statements of Social Responsibility and Good Practice Code of the ABB and RGA, all of which were in existence by the material time.

Applying the Barclays Bank test, Miss Day submitted that William Hill made a sufficient voluntary assumption of responsibility by the establishment and publication of its own Social Responsibility Policy with associated self-exclusion procedures, the mention of them in its Rules of Gambling, and by its internal procedures for identifying problem gamblers. As for the three stage test, she submitted that the inherent vulnerability of a problem gambler sufficiently satisfied the ‘reliance or dependence’ test and that it was reasonably foreseeable that a failure to apply the Social Responsibility Policy and Procedures in relation to a known problem gambler would risk causing that person both psychiatric and financial loss. The requisite proximity is established by the combination of the modest size and the vulnerability of problem gamblers as a class. There was she said nothing unfair in imposing a duty of care, since it went no further than the code of practice which William Hill had by then imposed upon itself, and for which it had trained its staff. Since the undertaking of a social responsibility towards problem gamblers was the quid pro quo (as she described it) for the ability of bookmakers to make large profits from gambling, there was nothing unfair in recognising a duty to take care in the application of that policy to problem gamblers so as to avoid unconsciously exploiting their vulnerability through carelessness. Finally, for the purposes of the incremental cross-check, Miss Day relied upon cases such as Watson v. BBB of C, Barrett v. MOD and Jebson v. MOD as showing that there were already established exceptions to the general principle that a person is not responsible for preventing his neighbour’s self-harm.

Miss Day’s narrower submission relied upon the considerations of foreseeability, proximity and fairness in much the same way as did her broad submission, but focused for the necessary assumption of responsibility upon the contents of the telephone conversation between the claimant and John on 5^th June 2006, which concluded with John’s assurance that:

“For the next six months … you will not be allowed to open [the account] under any circumstances. You will not be allowed to bet over the phone with William Hill.”

That assurance, she submitted, contained the requisite assumption of responsibility to do something, sufficient to import a legal duty to do it carefully. It was a narrow submission both because it was confined specifically to the creation of a special relationship between William Hill and the claimant, rather than with problem gamblers as a class, and because it had a narrow effect, namely simply to take care in providing a six month exclusion of the claimant from telephone gambling (by contrast for example with gambling at betting offices) with William Hill.

Attractively though it was put, I have not been persuaded by Miss Day’s broader submission. Its essential features were the identification of a duty of care to problem gamblers as a class, and a legal obligation to take care in the application of every aspect of William Hill’s Social Responsibility Policy and Procedures in respect of any customer reasonably identifiable as falling within that class.

I am not persuaded that by developing its own Social Responsibility Policy and Procedures William Hill can be said voluntarily to have assumed responsibility to all its problem gambler customers, in the sense of assuming responsibility to take care, with a concomitant liability to compensate customers injured in their mind or in their pocket by any failure to take care. My reasons follow.

First, I am not persuaded that it can properly be said of problem gamblers as a class that they uniformly suffer from such an impairment in their ability to control their own gambling that it would be proper for the law to treat them, without more, as being so vulnerable as to require special treatment, even in the absence of a request for it. The psychiatric evidence before me shows, (as it did in the Australian cases) that problem gamblers range across a wide spectrum from loss of control at the extreme end, to experiencing merely difficulties in controlling their gambling at the opposite end.

Secondly, while there is a sub-class of problem gamblers, namely pathological gamblers, for whom it can properly be said that control of their gambling has become impossible, it seems to me wholly unrealistic to suppose that in the ordinary course of its business a bookmaker can be expected to be able to identify that sub-class by way of what amounts to a process of medical diagnosis. Although in the present case I have been persuaded that the claimant had become a pathological gambler by the end of May 2006, it was no part of the claimant’s case that William Hill knew or ought to have known that he was within that sub-class at the extreme end of the more general class of problem gambler. Accordingly, while it may be that the broad duty of care alleged by Miss Day may arise in relation to a customer whose behaviour has become so extreme as to demonstrate to a bookmaker that his gambling is wholly outside his control, that possibility does not avail the claimant in the present case.

My third reason for rejecting the broad submission when tested by reference to the assumption of responsibility is that William Hill’s Social Responsibility Policy and Procedures were designed to be operated in such a way that any gambler offered self-exclusion would be required to acknowledge by his signature an express disclaimer of legal liability for the consequences of any gambling while excluded. This is a point which relates only to a claim for compensation for pure economic loss: see section 2(1) of the Unfair Contract Terms Act 1977. In principle, where A asks B to do something upon which A may rely (either because it is advice or because he is in a position of dependency) B is entitled to choose between three alternatives. He may decline altogether (and, unlike the Good Samaritan, pass by on the other side). He may do that which is requested with a disclaimer of responsibility for the consequences, or he may simply do it without qualification: see Hedley Byrne & Co v. Heller and Partners Ltd [1964] AC 465 at 486 per Lord Reid. In the present case William Hill’s Social Responsibility Policy and Procedures adopted the second of those three alternatives. The Gambling Act 2005 which was then shortly to come into force provided that compliance with licensing conditions designed (inter alia) to assist problem gamblers was not of itself to create legal liability, and the Gambling Commission’s own policy has, thus far, been to permit liability exclusions as compliant with the licensing regime. In this respect, it has followed a trend which appears to have been established in Australia and in two states in America, namely Pennsylvania and New York.

Turning to the three stage test, it seems to me again that the broad spectrum of differing levels of impairment of control of gambling falling within the general ‘problem gambler’ label impacts adversely at least at the second and third of those stages (i.e. proximity and fairness). Generally, it seems to me reasonably foreseeable that if a known problem gambler is permitted to continue gambling unrestrained, without an offer of self-exclusion or an invitation to seek counselling, he will be likely to suffer an aggravation of his condition due to the unrestrained feeding of his habit, and an ever growing risk of serious financial loss. But in my judgment the law should be very slow to recognise a sufficient proximity to justify a requirement to take protective steps to restrain a gambler from exercising his liberty to gamble on his own responsibility, where his status as a problem gambler may mean no more than that he is experiencing mild and occasional difficulties of control. Again, I emphasise that the broad submission advanced by Miss Day assumes a duty of care to all problem gambler customers, regardless whether they seek the bookmaker’s help. Such a duty would, in relation to a problem gambler who did not seek the bookmaker’s help, be an invasion of his autonomy, in relation to an activity for which he is primarily responsible for the consequences.

Nor does it seem to me that it would be fair to impose the broad duty of care for which Miss Day contends. As Mr Fenwick submitted, it would place a burden on the bookmaker pursuant to which the problem gambler could freely take home his profits, but look to the bookmaker for the return of his losses, without even seeking the bookmaker’s assistance to help him control his gambling.

Finally, the incremental cross-check seems to me to point away from, rather than towards, the recognition of a duty of care to problem gamblers by way of extension from established cases. For example, Watson v. BBB of C provides no real precedent because the bookmaker asserts nothing like the control over the customer’s activity which was exerted over the boxer in that case, or for that matter a control equivalent to that of a prison authority over a prisoner. Similarly, cases such as Barrett, Jebson and Munro v. Porthkerry Park Holiday Estates [1984] 81 LSG 1368 are decisions which go out of their way to emphasise the importance of the total incapacity of the intoxicated claimant, and a requirement that that be known to the defendant as having already occurred, or as being probable by reason of the circumstances.

I turn therefore to Miss Day’s narrower submissions, based upon John’s assurance to the claimant on 5^th June 2006. In this context it seems to me that the key features of the evidence are as follows. The claimant presented himself to John as a problem gambler, asked for help and was offered it in the form of a six months’ self-exclusion from gambling which he accepted. John then told the claimant that his account had been closed, and that for six months it would not be reopened, nor would he be permitted to gamble on the telephone with William Hill. The exchange was in all respects tantamount to a contract, save for the absence of consideration. In particular, the claimant was given to understand that he need do nothing more to obtain that self-excluded status. Because of John’s omission to process the claimant’s self-exclusion, the claimant was not required to sign a document clearly recording William Hill’s disclaimer of legal liability, or even his own obligation to comply with self-exclusion by abstinence. In summary therefore, the claimant presented himself as a problem gambler in need of help, asked for specific help, and was assured that he would be given it.

In my judgment those facts disclose a sufficient voluntary assumption of responsibility by William Hill to exclude the claimant from telephone gambling with the company for six months to give rise to a duty to take care to implement that exclusion, subject to the application of the three stage test, and the cross-check of incrementalism. In short, faced with a request for help from a person of some (albeit uncertain) vulnerability, William Hill chose to undertake to do that which was requested, without any disclaimer of legal responsibility.

Mr Fenwick submitted that since John’s undertaking to the claimant flowed out of a policy and procedure adopted by William Hill in preparation for the performance of a statutory obligation to comply with licensing conditions, it was not truly voluntary. I reject that submission. First, the policy and procedures were strictly voluntarily in every relevant sense in 2006, becoming part of a compulsory licensing regime a year later. More importantly, the fact that a person undertakes to do something because of a statutory obligation does not deprive that undertaking of its voluntary nature, for the reasons set out by Lords Hoffmann and Walker in the Barclays Bank case, to which I have already referred. The undertaking is voluntary because it is deliberate and intentional. At its highest the impending licensing regime may have made it in practice unrealistic for William Hill to choose the first of the three alternatives presented by the claimant’s request, namely to do nothing. But that left William Hill free to choose between the second and the third, and, albeit because of John’s mistake, the third (that is an undertaking unaccompanied by disclaimer) was chosen.

It may be asked why, if no voluntary assumption of responsibility to problem gamblers generally was incurred by William Hill by its adoption and promulgation of its Social Responsibility Policy and Procedures, should there be identified a voluntary assumption of responsibility merely because a problem gambler requests that the procedure be applied to him. In my judgment the request and the undertaking in response make all the difference. First, it brings the parties into a degree of relationship akin to that of contract, save only for the absence of consideration. Secondly, it deprives the objection that the provision of assistance may infringe the gambler’s autonomy of any force, since he has himself specifically requested it. As I have already suggested, the very essence of self-exclusion is that a problem gambler, recognising in a moment of clarity that he is likely to succumb to his addiction in the future, seeks his bookmaker’s assistance in helping him to control what he fears will be otherwise uncontrollable when temptation returns. He is, in effect, putting the bookmaker on notice of his fear that, at precisely the time when he wishes exclusion to be imposed upon him, he will himself be unable to control his gambling. Absent that fear, there would be no point in self-exclusion at all.

The assumption of responsibility test being satisfied on that narrower basis, I turn therefore to the three stage test. Again, it seems to me that both financial and psychiatric harm from a failure of the self-exclusion process is sufficiently foreseeable. There is a difficulty in relation to psychiatric harm that no physical harm was foreseeable, but Mr Fenwick placed no reliance upon this particular obstacle. In my judgment he was right not to do so. Problem gambling is a recognised psychiatric disorder from which the claimant was known by William Hill to be suffering. The case is far removed from cases such as White v Chief Constable. Although not on all fours with McLoughlin v Jones due to the absence of any contractual relationship, it comes within the general scope of the reasoning of Hale LJ at paragraphs 57 to 59.

Furthermore, there is no risk of indeterminate liability to an indeterminate class. In the context of a specific request for self-exclusion by a particular problem gambler, no question of class arises. Furthermore, the liability from a failure to implement a six month exclusion from telephone gambling with that bookmaker is limited in financial terms to losses over six months from telephone gambling, nothing more and nothing less. They are of course, losses wholly within the ability of the bookmaker to prevent, by implementing the self-exclusion process. As for psychiatric harm, the quantum likely to be available for an aggravated gambling disorder is so modest as to give rise to no real difficulty. As to fairness, I can see no particular reason why it should be regarded as unfair that a bookmaker who has undertaken, albeit without consideration, to exclude a problem gambler customer at his request, without making any disclaimer of liability, should incur a duty of care. Contrary to Mr Fenwick’s submission, he does not become the gambler’s loss insurer. His obligation is only to take reasonable care, and he has the concomitant right to prohibit the gambler from making any profit.

Appealing to a wider policy analysis, Mr Fenwick submitted that it would be wrong in principle to recognise a duty of care in relation to gambling losses, at a time before the repeal of the Gaming Act 1845, because it would appear to give rise to the possibility of recovery proceedings prohibited by the second limb of section 18. It would be, said Mr Fenwick, a suit … brought or maintained in a court of law and equity for recovering a sum of money deposited in the hands of a person to abide the event on which a wager was made. In that respect he relied upon the broad anti-avoidance construction given to the second limb of section 18 by the House of Lords in Hill v. William Hill (Park Lane) Ltd [1949] AC 530. Again, I reject that submission, for two reasons. The first is that section 18 of the Gaming Act 1845 was part of “legislation to discourage gaming in this country”: see per Viscount Simon at p.548. To construe section 18 as rendering void a self-exclusion arrangement designed to prohibit rather than encourage gaming between the relevant parties for a specified period would be to set the literal words of section 18 at war with their purposive construction. It follows that a claim for compensation for a failure to use reasonable care to implement a self-exclusion arrangement is not a suit to recover money deposited to abide the outcome of a wager within the meaning of section 18, merely because the financial ruin of the claimant for which compensation is sought has been caused by gaming losses during the specified period.

The second reason is that it would in my judgment be wrong to address the duty of care issue by reference to an outmoded policy which had been cast aside by the Gambling Act 2005 by the relevant time, merely because Parliament thought fit to provide a period of time before its coming into force, for the gaming industry to remodel its practices in line with the new underlying policy. The relationship between the claimant and William Hill in the context of which his request for self-exclusion arose was one substantially moulded by the new policy, even though it had yet to come fully into force in the form of mandatory licensing conditions.

Mr Fenwick made considerable play of the irony that, had John performed his role within the Social Responsibility Policy and Procedures correctly, William Hill would have disclaimed any relevant responsibility. In my judgment, the proper case for that argument lies at the causation stage of the analysis. Where a duty of care arises only because a party has not made a disclaimer which, had it been made, the law might have recognised, there is nothing unfair in requiring a party to take reasonable care in the implementation of that which it has undertaken to do without qualification.

Finally, in relation to the question of policy, I can envisage no floodgates being opened by recognition that William Hill assumed responsibility to the claimant in the particular circumstances of the 5^th June conversation. It arose from a combination of John’s assurance to the claimant that he need do nothing more to exclude himself, coupled with John’s own failure to process the claimant’s request in any way. It emerged only in the course of cross-examination and thereafter from documents belatedly disclosed that John has been disciplined for what he did and that, but for the fact that he was under considerable pressure of work at the time, his misconduct would have been regarded as very serious, and therefore unusual. In the ordinary run of cases, I have no reason to suppose that problem gamblers would either be promised self-exclusion in the course of an initial conversation, or given it in due course without a clear reliance upon William Hill’s standard disclaimer. Since the specimen self-exclusion agreement promulgated by the ABB also contains a disclaimer, I see no reason why the recognition of a duty of care in the special circumstances of this case should open floodgates in relation to the affairs of other bookmakers.

Turning finally to the cross-check of incrementalism, the cases relied upon by Miss Day as precedents for her broad submission offer no greater assistance for her narrow one. None of the claimants in those cases asked for, were offered but then failed to obtain the requisite protection or assistance. Nonetheless there are, as it seems to me, sufficient examples in the law generally where persons whose vulnerability impairs their decision making, but which falls short of depriving them of legal capacity, are treated as deserving special protection. Persons suffering from undue influence are an obvious example. Once that vulnerability is or should be apparent to persons such as solicitors advising them, or to persons such as banks seeking to deal with them at arm’s length, it is well recognised that special steps need to be taken.

Furthermore, I consider that the ‘all the indicia of contract save for consideration’ analysis of the dealings between the claimant and John on 5^th June means that the assumption of responsibility arising from their exchange can properly be extrapolated from the main line of authorities stemming from the Hedley Byrne case, even though the present case is more about a request for assistance by a person whose vulnerability gives rise to a degree of dependence, than a mere request for advice or information, the accuracy or truth of which is then to be relied upon. In the present case, the nature of the reliance was of course different, but it is in my judgment inherent in a request for self-exclusion that the problem gambler is seeking to rely upon the bookmaker’s assistance in maintaining his diminished control over his gambling, which, without that assistance, he fears will fail him.

I therefore decide the duty of care issue in favour of the claimant, but only on the narrower basis upon which it has been put forward.

In the light of the facts as I have described them, the breach of duty issue is straightforward. John’s failure to implement the self-exclusion arrangement which he had made on the telephone with the claimant was plainly careless, and it was not suggested that William Hill is otherwise than responsible for it.

A separate question arises whether William Hill’s self-exclusion procedures were defective to the extent that they constituted an additional aspect of breach of duty. I have recorded how, prior to 2007, the placing of a Vet Code 91 exclusion notice on a closed account would not of itself prevent the problem gambler from opening a fresh account, and I have described that as an apparent shortcoming in the procedure. In the claimant’s case however, because of his status as a greyhound trainer and his previous gambling success, he was closely monitored by Betting Control in relation to every bet placed after he opened Account No 2 on 5^th August. The evidence of Mr Richard Banks, William Hill’s Betting Control Director and Mr Michael Eckley, its Betting Control Manager, persuaded me that, had a Vet Code 91 note been placed on the claimant’s Account No 1 in June, Betting Control would have refused authority to accept any telephone bets from him thereafter, even if he had succeeded initially in opening a fresh telephone betting account. It follows that this structural defect, which was in 2007 remedied in the light of experience, in particular of the claimant’s case, constituted no breach of duty to him, even though it might have led to more modest self-excluded problem gamblers placing small bets which did not require Betting Control authorisation, during a period of self-exclusion.

Finally, my rejection of the wider basis upon which the claimant asserted a duty of care makes it unnecessary for me to deal with the allegations that there was a breach of duty in failing to refer the claimant for GamCare for counselling and advice, and a breach of duty by Elaine in May 2006 in the manner in which she handled the claimant’s requests to close his account permanently. In the event that a different view is taken by a higher court on the wider duty of care case, I have set out the relevant facts in full.

The first question is: what would have happened if John had duly implemented the claimant’s self-exclusion request, and William Hill had in other respects applied its Procedure, without any further lack of care? Mr Fenwick submitted that there were only two possibilities, both fatal to the claimant’s claim. The first was that he would have been offered a self-exclusion agreement by Customer Services on the following day and declined it. The second is that he would have been offered such an agreement, and accepted it, on terms by which William Hill disclaimed liability for any further consequences. In my judgment Mr Fenwick was right to propose those alternatives for decision, but wrong to stop the clock at the signing of a self-exclusion agreement.

It is possible that upon being presented with a self-exclusion agreement by Customer Services, the claimant would have changed his mind and declined to proceed with it further. There is certainly an air of impulse about the way he described his decision to speak to John, and there are two examples of occasions when, offered self-exclusion, the claimant declined it, in May and October 2006. Nonetheless, it seems to me on balance that having decided to self-exclude rather than merely close his account in June, the claimant would probably have proceeded to sign a self-exclusion agreement if presented with one by Customer Services on the following day, so that William Hill would (subject to any question as to its reasonableness) have been protected by its disclaimer for any subsequent carelessness. But the real question is whether, had he signed a self-exclusion agreement, he would have been excluded from gambling by William Hill thereafter, and if so for how long, and from what form of gambling.

In my judgment, had he signed a self-exclusion agreement on (say) 6^th June, the claimant would have been effectively excluded from telephone gambling for the following six months. That is because a Vet Code 91 notice would have been placed on his original account and, although it would not have prevented him from opening a new account, it would have prevented him from placing any significant bets, for the reasons which I have already explained.

I consider it on balance less than probable that, had the claimant signed a self-exclusion agreement, he would also have been prohibited by William Hill from gambling at its betting offices or, had he attempted it, from internet gambling with William Hill. The possibility arises because of the closeness with which his betting activity was monitored by Betting Control, whose jurisdiction necessarily extended to betting shops and all forms of gambling offered by William Hill, so as to provide effective protection against risk. But William Hill’s procedures required exclusion from gambling at betting offices to be separately arranged, and it did not follow from a Vet Code 91 notice being placed on a telephone betting account that the customer had also self excluded from betting offices, or even that he wished to do so. But in that respect his exclusion would in any event have gone beyond the scope of William Hill’s duty of care. On the narrow basis upon which I have found a duty of care to be incurred, it extended only to taking reasonable care to prevent the claimant from gambling with William Hill on the telephone. Furthermore, I am not at all persuaded that, if advised of the opportunity to do so, the claimant would have taken steps to exclude himself from betting offices, not least because (rightly or wrongly) he regarded telephone betting as exposing him to particular vulnerability by reason of its disassociation from the handling of hard cash.

It follows of course that the particular losses which the claimant sustained between August and December 2006 by reason of his telephone betting would not have been sustained, but for William Hill’s negligence. But that by no means concludes the causation analysis. Although in a sense the claimant’s case is that he was harmed by the aggregate outcome of the particular bets which he placed with William Hill, his complaint is that by failing to exclude him from gambling, William Hill caused his financial and social ruin and an aggravation of his gambling disorder.

However unsatisfactory this may be to philosophers and legal academics, causation is, as applied by the courts, ultimately a matter of common sense: see Galoo Ltd v. Bright Graham Murray [1994] 1 WLR 1360, applying dicta from Australia in Alexander v. Cambridge Credit Corporation Ltd (1987) 9 NSWLR 310 and March v. E & MH Stramare Pty Ltd (1991) 171 CLR 506. It would in my opinion fly in the face of common sense and be a travesty of justice if a problem gambler were able to attribute liability for his financial ruin to a particular bookmaker with whom he had made the relevant losses due to their failure to exclude him at his request, if he would, had he been excluded by that bookmaker, probably have ruined himself by betting with one or more of that bookmaker’s competitors. The position would of course be otherwise if the problem gambler had sought to exclude himself from betting at any bookmakers by separate arrangements with each, since it would be no answer by one bookmaker to a claim for compensation for negligence to say that, but for his negligence, the gambler would have been harmed in the same way by the negligence of another.

It follows that in my judgment it is essential for the court to form a view about what would have happened to the claimant’s gambling career if he had been excluded from telephone betting with William Hill. If the conclusion is that he would still have suffered financial and social ruin and a similar aggravation of his gambling disorder by betting with others, accompanied by more intensified gambling at William Hill’s betting offices, it seems to me that as a matter of common sense the claim must fail on causation grounds. William Hill’s negligence may have been a sine qua non for his particular gambling losses, but would not have been the effective cause of his ruin.

In this respect, the claimant’s approach was not to treat this further causation inquiry as irrelevant, but to address it by evidence suggesting that, had he been excluded from telephone gambling at William Hill, he would have not have been ruined by his addiction. I have described in my findings of fact the claimant’s own evidence to the effect that, apart from William Hill, he had either excluded himself or been excluded by the other major bookmakers and that, to the extent that he had not, the gambling opportunities still available to him from mid-2006 were so limited as to be of no sufficient attraction to him. As will be apparent, I found that aspect of the claimant’s evidence to be much less than satisfactory, in particular in relation to the critical period from June 2006 onwards.

The claimant’s solicitors sought to bolster his case in this respect by some discreet gambling enquiries of their own. The results were verified in a witness statement from Miss Tiejha Smyth, the solicitor with the day to day conduct of the case for the claimant, and she was cross-examined. Her evidence showed that she conducted a theoretical exercise in early 2008, directed to ascertaining the maximum bets which the claimant was likely to be permitted to make at betting shops in Tyne and Wear and County Durham, including Newcastle itself, in each case by seeking to place a cash bet on Sunderland to win in a football match against Birmingham. She received responses, depending on the odds, ranging between nil and £5,000. She also asked whether any of the local bookmakers offered a telephone betting service and received what would, had the claimant applied, have been a negative response.

I accept the accuracy of the evidence of what Miss Smyth ascertained, but the value of that evidence in resolving the causation issue which I have identified was severely limited by the restrictions upon her brief. She refrained from telephoning Coral or Ladbrokes on the basis of her client’s evidence that he had been excluded or severely limited by them. She refrained from any examination of the opportunities presented by internet betting, on the understandable basis that nothing other than the placing of real cash would have enabled her to obtain any useful information. Nor did she make any inquiry into the availability of telephone betting beyond the locality which I have described. The result is that her evidence by no means closed off the avenues by which, had he been excluded from telephone betting with William Hill, the claimant might otherwise have pursued his gambling, at a level and frequency sufficient to feed his addiction.

Each party criticised the other for failing to deploy satisfactory evidence in relation to this issue. Miss Day criticised William Hill for failing to adduce any evidence from bookmakers with whom the claimant said that it had become in practice impossible to place substantial bets by mid-2006. Mr Fenwick’s criticisms were focused upon the unsatisfactory aspects of the claimant’s own evidence, and the restricted scope of the evidence of Miss Smyth. In my judgment the reliability of Mr Calvert’s own evidence on this point was sufficiently undermined both by cross-examination and by the existence of apparently inconsistent documents, in particular in relation to the period after June 2006, that I should be very cautious before accepting his evidence on these matters in the absence of corroboration. Neither Miss Day nor Mr Fenwick sought to argue that this was a loss of a chance case.

My conclusions are as follows. First, the only bookmaker other than William Hill which has, after inquiry of all of them, confirmed that the claimant closed his account on the basis that it be not reopened is Stan James. Both Pagebet’s and Ladbrokes’ responses suggest otherwise. Secondly, I consider that the claimant exaggerated the degree of the restrictions which he claimed to have encountered in seeking to place bets otherwise than with William Hill after June 2006.

Thirdly I consider that, had he been excluded from telephone betting by William Hill in the second half of 2006, he would have sought other avenues for large scale betting, whether on the telephone or on the internet, and would have continued his gambling, albeit on a lesser daily scale than that in which he indulged with William Hill. Fourthly, I am satisfied on the balance of probabilities that, even if deprived of the opportunity to pursue his telephone gambling with William Hill in the second half of 2006, the claimant would ultimately have ruined himself financially, albeit at a slower rate because of the reduced scale of gambling which would have been available to him. The actual rate at which his alternative route would have led to his ruin must remain a matter of speculation, but it is sufficient for present purposes for me to conclude, as I do, that he would have ruined himself by the end of 2007.

I reach those conclusions for the following reasons. First, the claimant was already a pathological gambler before William Hill assumed, still less breached, any duty of care owed to him. Notwithstanding intermittent periods of clarity, he had in substance so lost control of his gambling that he was already on the road to ruin, once his luck ran out, as in due course it inevitably did. His scale of gambling by May 2006 was so large by comparison with his available resources that he could not endure any significant run of losses without running out of money.

Secondly, his gambling was driven not merely by the ‘buzz’ which he derived from it, but also by the fact that his personal and business lifestyle had come to depend to a substantial extent upon his gambling profits. I have in that respect rejected his evidence to the contrary.

Thirdly, his attempt at self-exclusion with William Hill in June 2006 did not of itself prevent him from succumbing again to the urge to gamble, for example through Betfair, before he even discovered in August that William Hill had not in fact excluded him. Nor did any imposition of restrictions by some of those bookmakers who he approached in the period immediately after June lead him to stop gambling, since it is clear that he continued to try other avenues.

Fourthly, and notwithstanding his denial in cross examination, I consider that the substantial round sum credits to and debits from his bank accounts during the period following June 2006 were probably receipts and payments in connection with further gambling which the claimant has chosen not to reveal. If, which I doubt, those receipts and payments were in connection with the sale and purchase of greyhounds, then the claimant can only have himself to blame for failing (for motives good or bad) to make any documentary record of those transactions.

Fifthly, there is real force in Mr Fenwick’s submission that there must have come a time on or after 5^th August 2006 when the claimant in a moment of clarity appreciated that William Hill had not in fact implemented any form of self-exclusion in relation to his telephone gambling, such that I ought to conclude that he was thereafter discharged from any false sense of security, occasioned by John’s assurance on the telephone on 5^th June. It is, again, a matter of speculation when the claimant had another moment or period of relative clarity during which this must have become apparent to him. In that respect the exchange which he had on the telephone on 10^th October is instructive. He chose not to go through a self-exclusion process after being told in unmistakable terms that if he did not, there would be nothing to stop him reopening his account the following week. In fact he did just that eight days later. The episode betrays a man with no commitment to enforced abstinence even in his moments of relative clarity. It is hard to see how, after that incident, his continuation on the road to ruin by gambling on the telephone can be laid at William Hill’s door.

A conclusion that William Hill’s negligence contributed to the claimant’s financial ruin only by accelerating what would probably have occurred in any event raises the question whether this process of acceleration caused any loss recognisable by the common law. I have asked myself whether for example a slower progression towards financial ruin would have resulted in the claimant borrowing less from his friends and acquaintances, or whether a slower progression to a point when further gambling became impossible would have lightened the burden of the psychological harm caused by his already pathological disorder.

As to the first, nothing in the evidence persuades me that a longer road to ruin would have reduced his net borrowings at the end of the day. The claimant’s disorder was such as to compel him to take any course short of outright crime to feed his habit, and in particular to conceal the reasons for his need for money from his lenders, and from his family. It is pure speculation whether a longer road to ruin would have led to his having been unmasked sooner, so as to cut off any further source of borrowed money.

I have reached the same conclusion in relation to psychological harm. At the worst, the claimant’s negligence enabled him to indulge his addiction at a rate which may well have aggravated its effect upon him. But the consequence was that it took only six months for him to be reduced to a position of enforced and, to date, permanent abstinence which, according to both psychiatric experts, appears so far to have released him from the craving to gamble. It seems to me entirely a matter of speculation whether a longer journey to the same destination would have been reflected in a net decrease rather than an increase in psychological harm and, no doubt because of the lateness of the amendment to plead a personal injury claim, no medical or other evidence was addressed to this issue anyway.

It follows that the claimant’s case entirely fails upon the ground that William Hill’s negligence merely affected the manner in which, and in particular the rate at which, a pre-existing pathological gambling disorder caused the financial and social ruin and the psychological harm which form the basis of his claim, without in any definable way increasing the aggregate amount of either form of harm.

In reaching that conclusion I have not ignored the submission that, by carelessly continuing to engage in gambling transactions with the claimant during the second half of 2006, William Hill unwittingly exploited his vulnerability. It is tempting to try and fashion a remedy directed to requiring the bookmaker in such circumstances to disgorge its profit, in particular by analogy with the remedies available in equity to the victim of undue influence, even as against third parties with the requisite knowledge.

There are in my judgment at least two insuperable difficulties with such an analysis. The first is that the bookmaker’s profit is not the obverse of the problem gambler’s loss, in particular where the gambler bets on the enormous scale indulged in by the claimant. Large bets cause a bookmaker to revise its odds, such that its true profit from the gambler’s losses requires a complex calculation by reference to all the other bets won and lost on the same event.

The second objection is that the common law of negligence seeks only to award compensation for loss, and does so by enquiring what the victim’s position would be if the relevant breach of duty had not occurred. The equitable remedy of an account of profits operates in accordance with different principles, and there is in my judgment no simple crossover, in the absence of any proprietary equity or relevant fiduciary duty, by which the two can be run together. For the purposes of the common law remedy, once it is ascertained that the claimant would have been no worse off if the relevant duty had been performed, that is an end of the matter.

In the circumstances, the question of quantum does not arise. In the event that a different analysis of the issue of causation might require it to be considered by a higher court, the relevant evidence as to the amount of the telephone betting, the amount of the losses and the amount of the interest incurred by the claimant by funding his habit is, subject to one point, both available and uncontentious.

The one exception lies in the extent and profit/loss consequences of the claimant’s gambling at betting offices during the second half of 2006. The agreed account of those transactions which can be identified with certainty suggests an overall loss of £136,000, but a documentary summary provided by William Hill to the claimant at his own request in 2006 suggested a net profit, by reference to other transactions of which no specific documentary record now survives. It is sufficient for present purposes for me to conclude, as I do, that the claimant’s gambling in William Hill’s betting offices during the second half of 2006 has not been shown to have caused him a net loss. I exclude from that analysis the cash deposited in betting shops to fund his telephone betting on his Account No 3, the outcome of which is precisely documented.

For the same reasons, the question of contributory negligence does not arise either. Contributory negligence in the case of self-inflicted harm where the defendant’s negligence has been an original or contributory cause was the subject of a division of opinion in the speeches in Corr v. IBC Vehicles Ltd [2008] UKHL 13, about which it is unnecessary for me to comment. Suffice it to say that if I had otherwise been persuaded that financial loss or psychiatric injury had been caused by the claimant’s negligence, the undoubted continued responsibility of the claimant in relation to his gambling would, however serious his loss of control during the second half of 2006, have led to a very large reduction of any award on the grounds of contributory negligence. That is first because, although his gambling was generally out of control during that period, the nature of the disorder is such that even its most serious sufferers have periods of relative clarity in which they could, if they wished, take steps not limited to self-exclusion, in order to try to deal with their habit. Secondly, I would also have regarded the claimant as having been mainly responsible for any loss which had been proved, because he was already a pathological gambler as a result of conduct for which he was entirely responsible before William Hill undertook, still less breached, any duty of care towards him.

Standing back for a moment to look at the wood rather than the trees, it is perhaps unsurprising that a negligent failure by a bookmaker in connection with the now officially sanctioned process of self-exclusion failed to cause the problem gambler any measurable loss. The conclusion flows in my judgment naturally from the inherently limited effectiveness of self-exclusion as a remedy for the underlying problem. In the form applied under William Hill’s Social Responsibility Policy, sanctioned by the ABB, the RGA and now by the Gambling Commission, the self-exclusion process is designed to operate privately as between a problem gambler and an individual bookmaker. There are no doubt powerful reasons of autonomy and privacy which prevent such a private two party arrangement being either known about, still less acted upon, by any of the individual bookmaker’s competitors. Since self-exclusion is a process entered into in recognition of the likelihood that the problem gambler will in future succumb to temptation without external assistance, it seems to me inherently likely that a gambler who has self-excluded with a single bookmaker will, when the temptation to gamble returns in force, simply go elsewhere. A telephone gambling or internet account with a new bookmaker takes a few moments to establish, and there is plenty of competition for the provision for those services.

The self-exclusion procedure forms a main plank in the social responsibility structure which stands as the quid per quo for the modern policy of the encouragement of gambling as an industry and as a leisure activity. The question whether the limited effectiveness of self-exclusion as a remedy for problem gambling undermines the integrity of that public policy bargain is something for the Gambling Commission and Parliament rather than the courts to decide.