some default text...
Judgments - SCA Packaging Limited (Appellants) v Boyle (Respondent) (Northern Ireland)

(back to preceding text)

58.   Despite four months of speech therapy and following this regime, the vocal nodes remained and in August 1992 she had another operation to remove them. After this she had further speech therapy and had to rest her voice completely for another four months. She carefully followed the management regime “which had a severe and upsetting effect upon her life” (para 8(7) of the Industrial Tribunal’s decision). When she returned to work she had to stagger telephone calls, limit the number and length of meetings, rest her voice during breaks, control the temperature in her office, plan the use of her voice, speak quietly and not shout, and not compete with background noise. The vocal nodes did not recur after 1992. The applicant put this down to her strict adherence to the management regime, with the severe curtailment of her social and leisure activities which this entailed. The respondent maintained that she had been cured.

59.  Matters came to a head in September 2000 when her line management changed. Her new manager decided to take down the partition separating her office from the stock control room. She thought that the increased noise levels would have a substantial adverse effect upon her health. She complained, with support of her ENT surgeon, but the employer’s occupational health specialist took a different view. In October 2001, she launched proceedings under the 1995 Act, complaining of discrimination. This must have been because of the employer’s failure to make reasonable adjustments to cater for her disability, contrary to what is now section 3A(2) of the 1995 Act. The employers denied discrimination and claimed that their proposals were fair and reasonable and made for justifiable operational reasons. They also denied that she suffered from a disability within the meaning of the Act. In February 2002 she was told that her position was to be made redundant and brought further proceedings alleging discrimination and/or victimisation. In May 2002 she was made redundant and after an unsuccessful appeal brought further proceedings in November 2002 under the 1995 Act and for sex discrimination and unfair dismissal. All these proceedings were eventually consolidated but in November 2004 it was directed that there should be a preliminary hearing on whether the applicant had a disability within the meaning of the 1995 Act.

60.  That hearing eventually began in November 2005 and took place over five days ending in February 2006. The Tribunal heard evidence from the applicant’s speech therapist, three consultant ENT surgeons and a specialist in occupational health. The decision was issued in May 2006. The Tribunal (Mrs O Murray, sitting alone) found that the applicant did suffer from a physical impairment in the form of hoarseness and vocal nodes (both of them listed in the WHO classification of diseases). She found that when affected by vocal nodes and their aftermath, the applicant suffered an adverse effect upon her normal day to day activities - the ability to talk without losing one’s voice or vocal volume, to converse without having to plan voice-use and without having to allow for voice rest after moderate use, to talk on the telephone without having to take compensatory lengthy voice rest, and so on. She found that the effect upon the applicant when she suffered from the effect of vocal nodes was substantial. She found that the management regime did constitute “treatment” within the meaning of paragraph 6(1) of Schedule 1 (paragraph 48 above). Ignoring its effects, she found that the applicant “would have suffered from the hoarseness and ultimately the nodules” during the relevant period had she not followed the regime: para 31. The management regime constituted a great curtailment of day to day activities and went far beyond the reasonable “coping strategies” envisaged in the Departmental Guidance: Guidance on matters to be taken into account in determining questions relating to the definition of disability, Department of Economic Development (1996). Hence she found that, under paragraph 6(1), the substantial adverse effect was “deemed to have continued” throughout the period: para 36. However, she also found that, on the balance of probabilities, the condition of vocal nodules was likely to recur for the purpose of paragraph 2(2). The doctors were agreed that the applicant had a propensity to develop them. The only reason she had not done so was her very strict adherence to the management regime, “to an extent which went far beyond any reasonable measures to be taken to preserve voice quality or avoid vocal overuse, misuse or abuse": para 41. Hence the Tribunal found that both paragraphs 6(1) and 2(2) of Schedule 1 to the 1995 Act (paragraphs 48 and 49 above) applied and the applicant was a disabled person within the meaning of the Act.

61.  It took until February 2008 for this preliminary issue to reach the Court of Appeal. Much of this delay was attributable to the process of stating a case for the opinion of the Court of Appeal which, as my noble and learned friend Lord Hope of Craighead has explained, continues to apply in Northern Ireland. The Court of Appeal gave judgment in October 2008 dismissing the employers’ appeal. The Court accepted that the Tribunal’s decision was open to the criticism that paragraphs 2(2) and 6(1) of the Schedule are concerned with the recurrence or correction of the adverse effects of an existing impairment. It is still necessary for an impairment to exist. But the Court felt it possible to draw that inference from the findings which the Tribunal had made.

62.  More importantly for our purposes, the Court of Appeal discussed what was meant by the words “likely to have a substantial adverse effect” in paragraph 6(1) and concluded that “likely” was used in the sense of “could well happen” rather than probable or more likely than not.

The arguments

63.  Noelle McGrenera QC, for the appellant employers, argues that this interpretation is wrong. Had the Court of Appeal applied the correct interpretation, it would have been bound to find that the decision of the Tribunal had been perverse on the evidence presented to it. Robin Allen QC, for the interveners, holds no brief for the respondent employee. But he points out that all the indications are that the Tribunal applied the higher test of likelihood when applying paragraphs 2(2) and 6(1) of Schedule 1. He also suggests that the Tribunal was entitled to reach the conclusions which she did on the evidence before her. Thus in any event the appeal should fail.

64.  Further, he makes the important point that, if the substantial adverse effects of an impairment are being held at bay by a course of treatment or other corrective measures, so that paragraph 6(1) applies, there is no room to consider the likelihood of recurrence under paragraph 2(2). Paragraph 2(2) assumes that the substantial adverse effects have gone away for the time being without treatment and the question is whether they are likely to recur at some point in the future. Under paragraph 6(1) the question is whether the effect would be likely were it not for the corrective treatment. He also submits that the meaning of “likely” in both paragraphs is that given by the Northern Ireland Court of Appeal.


65.  The Northern Ireland Department of Economic Development has issued Guidance on the matters to be taken into account in determining questions relating to the definition of disability, under section 3 of the 1995 Act. The Guidance current at the time was issued in 1996 and is to the same effect as the Guidance issued by the Secretary of State for England and Wales (both have since been replaced). Both state, at para B7, that “It is likely that an event will happen if it is more probable than not that it will happen". This curious statement appears to have got “likely” and “probable” the wrong way round. It is probable that an event will happen if it is more likely than not that it will do so. Probability denotes a degree of likelihood greater than 50%. Likelihood, on the other hand, is a much more variable concept.

66.  Nevertheless, the English case law has until now adopted this approach. In Latchman v Reed Business Information Ltd [2002] ICR 1453, the Employment Appeal Tribunal concluded that an effect was not “likely” to recur if the risk of recurrence was about 50%. That test was simply repeated by the EAT in Swift v Chief Constable of Wiltshire Constabulary [2004] ICR 909, at para 28, referring both to Latchman and to the Guidance. There is a more extended discussion in Eastern and Coastal Kent PCT v Grey, UKEAT/0454/08/RN, at paras 18 to 23, where the EAT considered that it meant something other than “may” or “might” (para 22) but nevertheless concluded, somewhat mysteriously, that the use of the word “would” by the Tribunal showed that a higher threshold had been reached than was required by the word “likely” (para 23). Not only that, in Cunningham v Ballylaw Foods Limited [2007] NICA 7, para 11, the Northern Ireland Court of Appeal also cited para B7 of the Guidance with apparent approval. So it would appear that, until this case, the Guidance has met with an uncritical response on both sides of the Irish sea, although only in Latchman was it decisive of the outcome.

67.  In this House, we start with a clean slate. The Guidance has, of course, to be taken seriously into account when it deals with the factual matters which are relevant to the application of the legal tests. It is common for statutory Guidance to try to explain, not only how the legislation should be put into effect by the people who have to apply it, but also what the legislation means. But that is simply being helpful to practitioners who are not lawyers and may never read the legal texts. Statutory construction remains a matter for the courts, not for Departmental Guidance. If the court considers that the Guidance is a mis-statement or mis-application of what Parliament has enacted, then it must say so.

68.  It is significant that, apart from the EAT decisions cited above, our attention has been drawn to no case in which “likely” has been held to mean “more likely than not". This is scarcely surprising, as Parliament can always use the word “probable” if that is what it means. In Cream Holdings Ltd v Banerjee [2004] UKHL 44, [2005] 1 AC 253, Lord Nicholls of Birkenhead said this, at para 12:

“As with most ordinary English words “likely” has several different shades of meaning. Its meaning depends upon the context in which it is being used. Even when read in context its meaning is not always precise. It is capable of encompassing different degrees of likelihood, varying from ‘more likely than not’ to ‘may well'.”

In that case, the House related the degree of likelihood of success at trial, required by section 12(3) of the Human Rights Act 1998, to the potential adverse effects of disclosure. This is very similar to the position in child care cases, where the degree of likelihood of harm required by section 31(2) of the Children Act 1989 is related to the seriousness of the consequences if nothing is done. As Lord Nicholls said in In re H (Minors)(Sexual Abuse: Standard of Proof) [1996] AC 563, at p 585:

“The context shows that ... likely is being used in the sense of a real possibility, a possibility that cannot sensibly be ignored having regard to the nature and gravity of the feared harm in the particular case".

69.  There are very good reasons for concluding that, in this case too, Parliament did not intend that “likely” should mean “more likely than not". We are used, in civil proceedings, to deciding whether or not something has happened in the past “on the balance of probabilities". We ask ourselves whether it is more likely than not that something happened. We usually have a good deal of evidence to help us decide what went on. Once we have done so the event is treated as a fact: it was probable, therefore it was certain: see, for example, In re B (Children)(Care Proceedings: Standard of Proof)(CAFCASS intervening) [2008] UKHL 35, [2009] 1 AC 11, per Lord Hoffmann, at para 2.

70.  But predictions are very different from findings of past fact. It is not a question of weighing the evidence and deciding whom to believe. It is a question of taking a large number of different predictive factors into account. There are cases, as my noble and learned friend Lord Rodger of Earlsferry points out, in which the doctors can predict with all too much confidence what will happen to the patient. But in many others, putting numbers on what may happen in the future is a guessing game. Who can say whether something is more than a 50/50 chance? That is what the doctor in Latchman found so difficult. But assessing whether something is a risk against which sensible precautions should be taken is an exercise we carry out all the time. As Girvan LJ put it in the Court of Appeal, at para 19:

“The prediction of medical outcomes is something which is frequently difficult. There are many quiescent conditions which are subject to medical treatment or drug regimes and which can give rise to serious consequences if the treatment or the drugs are stopped. These serious consequences may not inevitably happen and in any given case it may be impossible to say whether it is more probable than not that this will occur. This being so, it seems highly likely that in the context of paragraph 6(1) in the disability legislation the word “likely” is used in the sense of “could well happen".

It has often been emphasised in the cases that the burden of proving disability rests with the applicant, who must bring medical evidence to establish this. Witnesses from any branch of medicine (including the professions related to medicine such as speech therapy) will be far more comfortable with assessing the reality of the risk rather than putting precise percentages upon it.

71.   Furthermore, as Mr Allen points out, the finding of disability is a threshold. In most cases, the question is whether the employer should have made reasonable adjustments to cater for the disability. The real issue in this case is whether it was reasonable to expect the employer to continue to adjust the working environment to take account of Mrs Boyle’s problems with her voice or whether it was not. The employer needs to know this in real time, and not to have to wait until a Tribunal has heard all the evidence and reached a conclusion about what is more likely than not to happen in the future. As with the child care cases, the question here should be, are these adverse effects sufficiently likely to require us to consider what, if any, adjustment should be made to take account of them?

72.  We do not know the answer to that question in this case because the substance of the complaints has not yet been tried. It could be that, weighing the extent of Mrs Boyle’s problems against the dictates of the employer’s business, it was not reasonable to expect the employer to make adjustments for her. It could be that it was. The same evidence which was heard on the disability issue will also be relevant to the adjustment issue. It is most unfortunate that they have been separated in this case and that it has taken so long for the preliminary issue to be resolved.


73.  I therefore conclude that the Court of Appeal in Northern Ireland applied the right test. Latchman should be overruled and dicta to the same effect disapproved. That being so, there is no need for us to go on to consider whether the Tribunal’s findings of fact were perverse in the light of the evidence before her. But I would have been very reluctant to do so. There was clear evidence of an underlying propensity to develop serious problems with the voice. The person who was closest to the regime which the patient was following was the speech therapist, whose evidence was that failure to follow the regime would lead to deterioration in voice and recurrence of the vocal nodules. The ENT surgeons found it more difficult to be so confident but that does not mean that the Tribunal should not have reached the conclusion which she did.

74.  However, the fact that the surgeons found it so difficult to put precise numbers on their predictions, when all were agreed on the underlying propensity to voice problems, supports the view that this was not the exercise which Parliament expected of them.

75.  For these reasons I would dismiss this appeal. The case must now go back to the Tribunal for the substantive complaints to be determined unless, of course, the parties are able to reach agreement upon them. I would also agree with your lordships that, where a tribunal has issued a fully reasoned decision, the case stated procedure has “nothing whatever to commend it". It is suitable for appeals on points of law from courts or tribunals which do not routinely explain their decisions. But aside from the duplication of effort and delay, it can give rise to unseemly debates between the tribunal and the parties as to the issues upon which a case should be stated. It is the appeal court, rather than the tribunal under appeal, which should decide which issues are worthy of its attention. It is to be hoped, therefore, that the relevant rule-making authorities in Northern Ireland will consider making rules to provide for an ordinary appeal procedure in these cases.


My Lords,

76.  I have had the advantage of reading in draft the opinions of my noble and learned friends Lord Hope of Craighead, Lord Rodger of Earlsferry, Baroness Hale of Richmond and Lord Neuberger of Abbotsbury. I am in substantial agreement with all of them and in common with each of your Lordships I too would dismiss this appeal. I add only a few paragraphs of my own.

77.  If someone is a “disabled person” within the meaning of the Disability Discrimination Act 1995 others must in certain defined circumstances “make reasonable adjustments” in relation to him. If not disabled, however, the person can look to no such consideration. Take this respondent. No one disputes that during the relevant period, and indeed for some years beforehand, she had on medical advice submitted herself to a strict management regime in order to save her voice and prevent nodules recurring on her vocal cords. (Others of your Lordships have fully recounted the facts of this appeal and no purpose whatever would be served by my restating them, or indeed the governing legislation, again.) If disabled, she could look to the appellants, her employers, to “make reasonable adjustments” in determining the particular circumstances and conditions of her employment. For example, she might reasonably have been entitled to expect the appellants to take note of her voice management regime and not, say, subject her to changed circumstances requiring her to shout, or engage in long or frequent telephone calls, or work in a smoky or otherwise polluted atmosphere, or whatever else. If, however, she was not to be regarded as disabled, her employers would be free of these constraints.

78.  It is in this context that your Lordships are called upon to decide the true meaning and application of the word “likely” where it appears in paragraph 6(1) of Schedule 1 to the 1995 Act—or for that matter in paragraph 2(2) of the Schedule although, as Lady Hale helpfully points out, that paragraph serves a quite different purpose. Assume a serious risk exists that, but for an employee’s observance of whatever measures are being taken to treat or correct an impairment (in this case the management regime designed to combat the respondent’s propensity to develop vocal nodules), the substantial adverse effects of that impairment would recur, is it really to be said that, unless the risk can be shown to amount actually to a probability, the employer (subject only to ordinary employment law considerations) can simply ignore the employee’s condition and take no steps whatever, however ostensibly reasonable, to accommodate the employee’s need to continue the treatment measures? To my mind, plainly not. It is sufficient to establish that, were the treatment regime to be materially disrupted, the severe disabling effects could well recur. Such a finding would carry with it a requirement of cooperation on the employer’s part. So much for the substantive issue for decision.

79.  As for the procedural questions thrown up by the long and tortuous (not to say ultimately unproductive) course thus far taken in this dispute, there is little, save emphasis, that I wish to add to what others of your Lordships have already said. First, unless there is a probability (I use the word advisedly) that a preliminary issue as to whether the complainant is disabled or not will be determinative one way or the other of the entire dispute, it is highly unlikely to be justifiable: there will almost certainly be more to lose than to gain by such a process. Secondly, where, as here, the tribunal has issued a fully reasoned decision on the point at issue, the case stated procedure, as opposed to a straightforward appeal by leave, has nothing whatever to commend it—and much by way of needless delay, expense and general aggravation in its disfavour. It really is time to take a close look at the relevant rules with a view to eradicating this absurdity for the future.


My Lords,

80.  I have had the benefit of reading in draft the opinions of my noble and learned friends, Lord Hope of Craighead, Lord Rodger of Earlsferry and Baroness Hale of Richmond.

81.  There is nothing that I can add to Baroness Hale’s analysis of the meaning, effect and interrelationship of the various provisions of Schedule 1 to the Disability Discrimination Act 1995 which fall to be considered in this case, or to her explanation as to the meaning of the word “likely” in paragraphs 2(2) and 6(1) of that Schedule. For the reasons which she gives, and all of those given by Lord Rodger of Earlsferry I would dismiss this appeal.

Continue  Previous