The Information Commissioner v Niebel (Information rights : Data protection) [2014] UKUT 255 (AAC)

UKUT-AAC
The Information Commissioner v Niebel (Information rights : Data protection)
[2014] UKUT 255 (AAC) · 2014-06-11
[46]I accept the thrust of Mr Hopkins’s starting submission, namely that, absent some binding statutory definitions or jurisprudence, comprehensive definitions of ordinary English words are neither necessary nor helpful. Mr Hopkins referred me to the well-known observations of Lord Reid in Brutus v Cozens [1973] AC 854 to the effect that "The meaning of an ordinary word of the English language is not a question of law. The proper construction of a statute is a question of law” (at 861C; see also Lord Kilbrandon at 867B-C). Mr Hopkins also reminded me of the comments of Lord Hoffmann in Moyna v Secretary of State for Work and Pensions [2003] UKHL 44 ; [2003] 1 WLR 1929 , who noted that “many words or phrases are linguistically irreducible in the sense that any attempt to elucidate a sentence by replacing them with synonyms will change rather than explain its meaning” (at paragraph 23). 47. I interpose here that Lord Hoffmann’s warning echoed the observations of Lord Upjohn in Customs and Excise Commissioners v Top Ten Promotions Ltd [1969] 1 WLR 1163 (at 1171):
“It is highly dangerous, if not impossible, to attempt to place an accurate definition upon a word in common use; you can look up examples of its many uses if you want to in the Oxford Dictionary but that does not help on definition; in fact it probably only shows that the word normally defies definition. The task of the court in construing statutory language such as that which is before your Lordships is to look at the mischief at which the Act is directed and then, in that light, to consider whether as a matter of common sense and every day usage the known, proved or admitted or properly inferred facts of the particular case bring the case within the ordinary meaning of the words used by Parliament.” 48. That passage in Lord Upjohn’s opinion was cited recently by Upper Tribunal Judge Jacobs in AS v Secretary of State for Work and Pensions (ESA) [2013] UKUT 587 (AAC) where the issue was the meaning of the terms “repeatedly … within a reasonable timescale” in a test of entitlement to a social security benefit. Judge Jacobs held as follows: “19. I am not going to attempt to define what these words mean. That would be wrong. It would be the wrong approach to statutory interpretation and would trespass impermissibly into the role of the First-tier Tribunal. It is not for the Upper Tribunal to give more specific content to the law than the language used in the legislation. The Upper Tribunal will not decide that ‘repeatedly’ means five times, ten times or any other number. Nor will the Upper Tribunal decide that ‘reasonable timescale’ means five seconds, five minutes or any other time.”
[49]Although the context is very different, the same principle holds good here. With that very significant qualification in mind, I accept Mr Cornwell’s point that “substantial” can carry two rather different meanings, but the statutory language and context provides the clue. I have derived considerable assistance, despite the very different context, from Majorstake Ltd v Curtis [2008] UKHL 10 ; [2008] 1 AC 787 , a decision of the House of Lords on the interpretation of section 47(2)(b)(ii) of the Leasehold Reform, Housing and Urban Development Act 1993. That provision is concerned with whether a landlord intends to “ carry out substantial works of construction on the whole or a substantial part of the premises in which the flat is contained”. “Substantial” is used in two slightly different senses here – as an adjective describing the nature of something (“substantial works”) and as an adjective defining a part of a defined or known greater whole (“a substantial part of the premises”). Against that background, Lord Scott of Foscote held as follows (at paragraph 18):[18]“In the expression ‘substantial works’ the adjective ‘substantial’ denotes, in my opinion, works that are not trivial or, as one might say, insubstantial. There is no other yardstick than impression. The issue is one of fact and degree. The same approach should, in my opinion, be taken to the question whether Flats 74 and 77 constitute a ‘substantial’ part of Block B. ” 50. Baroness Hale of Richmond, giving what was in effect the leading judgment, held as follows (at paragraph 40):
"‘Substantial, is a word which has a wide range of meanings. Sometimes it can mean ‘not little’. Sometimes it can mean ‘almost complete’, as in ‘in substantial agreement’. Often it means ‘big’ or ‘solid’, as in a ‘substantial house’. Sometimes it means ‘weighty’ or ‘serious’, as in a "substantial reason"
. It will take its meaning from its context. But in an expression such as a ‘substantial part’ there is clearly an element of comparison with the whole: it is something other than a small or insignificant or insubstantial part. There may be both a qualitative element of size, weight or importance in its own right; and a quantitative element, of size, weight or importance in relation to the whole. The works intended by this landlord are substantial in relation to each of the flats involved, but those flats do not in my view constitute a substantial part of the whole premises. ” 51. In the present context of section 55A(1)(b), “substantial” is used as a descriptive qualification of “damage” (and indeed of “distress”). The guidance states that “t he Commissioner will consider whether the damage or distress is merely perceived or of real substance” (at p.14). The tribunal gave no indication that it dissented from that proposition. Rather, it recorded (at paragraph [11] of its decision) what it understood to be the agreed position as between the advocates that “substantial” could have both a quantitative and qualitative dimension but was ultimately a question of fact and degree. It does not seem to me that the tribunal can reasonably have been expected to deconstruct the term “substantial” any further. 52. Turning to the question of what amounts to “damage”, the Commissioner’s guidance suggests that “Damage is any financially quantifiable loss such as loss of profit or earnings, or other things” (see above at paragraph 14). Again, the tribunal did not take issue with that guidance. In the course of oral argument, Mr Cornwell invited me to adopt a broader approach, given the preliminary view of Tugendhat J in Vidal-Hall v Google Inc [2014] EWHC 13 (QB) (discussed further below) that damage within section 13 of the DPA includes non-pecuniary damage (at paragraph [103]), given European developments as to the concept of “moral damage”. 53. I decline that invitation. The legislation in issue here refers disjunctively to “ substantial damage or substantial distress”. Of course, that same phrase also occurs in section 10 of the DPA, but neither counsel referred me to any authorities on that statutory provision which might lend any assistance. It seems to me, given the distinction drawn by the statutory language in section 55A(1)(b), that the Commissioner’s guidance is right to differentiate between “damage” and “distress” in the way that it does. If “damage” was meant to encompass emotional turmoil, then there would have been no need to refer separately to “distress”. 54. The tribunal also adopted that same understanding. It is plain from paragraphs [26] and [27] of its decision that it was assessing the extent of what one might call those hard-edged economic losses caused by Mr Niebel’s activities, to see if they amounted to a likelihood of “substantial damage”. These included the modest costs for those who texted back “stop” (where they had already exceeded their text limit, if relevant) or who replied from abroad, as well as the opportunity costs involved. This was ultimately a classic jury fact-finding question. The tribunal concluded that in the context of the 286 unwanted text messages “it would be most unlikely for a contravention of the nature and scale described in para 24 above to cause substantial damage”. That finding is, in my assessment, simply unassailable. The tribunal applied the correct legal test and explained its reasoning adequately albeit concisely. This case was, moreover, a long way from the example of “substantial damage” given in the Commissioner’s own guidance (see paragraph 14 above). If the Commissioner was going to make good his case, it was realistically always going to have to be on the basis of a finding that there was a likelihood of “substantial distress” rather than “substantial damage”. Ground 3: the tribunal’s approach to “substantial distress” The First-tier Tribunal’s decision 55. The tribunal had this to say as to the meaning of “substantial distress”:
“[12] Finally, Mr Cornwell asked us to adopt a definition of distress produced by the ICO. This describes it as ‘any injury to feeling, harm or anxiety suffered by an individual’. We doubt the wisdom, for the purposes of applying the statute, in dividing up the phrase ‘substantial distress’ into two words. If the ICO definition involves the proposition that it is not possible to have ‘any injury to feelings’ which falls short of ‘distress’ then, it seems to us, that the definition is at odds with common experience and with the ordinary use of English.” 56. The tribunal later concluded as follows: “ F. Substantial distress [28] The ICO suggests that recipients of the accident claim texts might become concerned for the safety of members of their family or be disturbed by being reminded of a previous accident. Having looked at the wording of the texts, we judge this to be highly unlikely. Almost all mobile phone users, in our judgement, will recognise these texts for what they are. We also regard it as highly unlikely that the texts would evoke distress by raising false expectations of compensation. [29] In our judgement the effect of the contravention is likely to be widespread irritation but not widespread distress. Given the scale of the contravention, there is the possibility of some distress in very unusual circumstances but we cannot construct a logical likelihood of substantial distress as a result of the contravention. We conclude that the contravention is not of a kind likely to cause substantial distress.”
The parties’ submissions summarised 57. Mr Cornwell submitted that the tribunal erred in law in its approach to the meaning and application of “substantial distress” in five respects. First, the tribunal had failed to explain its interpretation of “substantial”. Second, the tribunal had failed to explain its interpretation of “distress”. Third, the tribunal had adopted an “unduly demanding” interpretation of “distress”. Fourth, the tribunal had failed to take into account the evidence before it of complainants’ distress. Fifth, the tribunal had substituted its own view of the impact of marketing texts for that of the actual recipients. 58. Mr Hopkins took issue with each and every one of these submissions. I need not rehearse his own arguments in detail here as, for the most part at least, I adopt them for the purposes of my own analysis. The Upper Tribunal’s analysis 59. Mr Cornwell’s first two points can be taken together. Having dismissed the likelihood of “substantial damage”, the question for the tribunal was whether there was, in the alternative, a likelihood of “substantial distress”. As noted above, “substantial” here is a qualifying adjective of the noun “distress”, so there is considerable force in the tribunal’s doubts about dividing the expression “substantial distress” into its component elements and then subjecting each to intense forensic scrutiny. The expression may indeed be “linguistically irreducible” in the terms used by Lord Hoffmann in Moyna . 60. That said, the tribunal took issue with the Commissioner’s guidance as to the meaning of “distress” and, in my opinion rightly so. According to that guidance, “Distress is any injury to feelings, harm or anxiety suffered by an individual” (at paragraph [12], emphasis added). The tribunal’s conclusion was that if this “ involves the proposition that it is not possible to have ‘any injury to feelings’ which falls short of ‘distress’ then, it seems to us, that the definition is at odds with common experience and with the ordinary use of English.” This was reinforced by the tribunal drawing a distinction between “irritation” and “distress” (at paragraph [29]); again, as a matter of ordinary English usage, that must be right. Thus the tribunal managed to explain the sense in which they understood “substantial distress” but without falling into the trap of seeking to replace the statutory test with a synonym which might carry different nuances of meaning. 61. Mr Cornwell’s third submission under this ground of appeal was that in doing so the tribunal had adopted an “unduly demanding” interpretation of the word “distress”. He sought to buttress this submission in four ways.[62]First, Mr Cornwell argued that the Court of Appeal has held that frustration experienced by a data subject by breach of EU data protection regulations is sufficient to constitute distress (see Halliday v Creation Consumer Finance Ltd [2013] EWCA Civ 333 ; [2013] 2 Info LR 85), a case involving a breach of the DPA by a finance company when processing personal data. That was an action under section 13 of the DPA, under which an individual who suffers damage (or distress) by reason of a data controller’s contravention of any of the Act’s requirements is entitled to compensation from the data controller for that damage (or distress). Mr Cornwell relied on paragraph 35 of Arden LJ’s judgment. Her Ladyship, having noted that there was “ no contemporary evidence of any manifestation of injury to feelings and distress apart from what one would normally expect from frustration at these prolonged and protracted events ”, then accepted “as a general principle that, where an important European instrument such as data protection has not been complied with, there ought to be an award, and it is to be expected that the complainant will be frustrated by the non-compliance.” It does not seem to me that this even begins to get Mr Cornwell’s argument home. Arden LJ’s observations were in the context of section 13, which refer to ‘damage’ or ‘distress’ simpliciter . The test under section 55A (as under section 10) is the appreciably different formulation of “substantial damage or substantial distress”. As Mr Hopkins put it, the addition of the epithet “substantial” raised the bar; the word must be there to do some work.63. Second, Mr Cornwell noted that the High Court has recently suggested that “damage” within section 13 of the DPA could be read to include “moral damage”, connoting the right to compensation for breach of individual rights even where the rights are non-pecuniary ( Vidal-Hall v Google Inc [2014] EWHC 13 (QB) per Tugendhat J at paragraphs [97] and [106]). This is far too slender a branch on which to hang such a weighty argument. With respect, Tugendhat J decided nothing beyond that there was a serious issue to be tried in that litigation; the observations on section 13 were therefore necessarily tentative. Furthermore, as with Halliday v Creation Consumer Finance Ltd , this decision concerned a provision in the DPA which uses a different formulation to section 55A(1)(b). As Mr Hopkins submits, Tugendhat J’s reasoning appears to collapse the distinction between ‘damage’ and ‘distress’. There is some support for Mr Hopkins in the Court of Appeal’s decision in Johnson v Medical Defence Union [2007] EWCA Civ 262 ; [2011] Info LR 110, where Buxton LJ expressed the view (admittedly obiter) that “there is no compelling reason to think that ‘damage’ in the Directive has to go beyond its root meaning of pecuniary loss” (at paragraph [74]). Whether or not Tugendhat J’s approach is a proper reading under section 13 of the DPA (in which the terms are deployed separately in subsections (1) and (2) respectively, which may not rule out some degree of overlap), in my judgment it cannot apply to section 55A(1)(b), where the expression “substantial damage or substantial distress” is plainly disjunctive.64. Third, Mr Cornwell submitted that the tribunal’s approach was inconsistent with recital (69) to the 2009 Directive, which refers to the need for effective implementation and enforcement powers. Thus the new Article 15a(1) of the 2002 Directive requires that Member States ensure that “the penalties provided for must be effective, proportionate and dissuasive”. The tribunal’s construction, he argued, undermined the stated legislative goal of protecting the privacy of individuals from aggressive electronic direct marketing techniques. However, as Mr Hopkins submits, this was a “bewitching submission” which took the Commissioner’s case no further. We can all agree on the need for a purposive interpretation, but this is an attempt to argue from a broad high-level policy objective to determine the outcome of a particular case. The tribunal’s task was to decide if there was a likelihood, on the evidence as it was presented, of “substantial damage or substantial distress”.65. As already noted, the formulation of the threshold in those terms was not mandated by EU law; it was a domestic decision to set the bar at that level. If the result of this case is a question mark over whether the Commissioner’s powers are in the terms of the Directive “effective” and “dissuasive” (howsoever those terms are defined), then it seems to me there are at least two possible answers. The first is for the Commissioner to present a more compelling case. Given the considerable resource and effort that was put into this investigation by the Commissioner’s staff, I suspect it is questionable whether much more could be achieved on that front. The second, and realistically more profitable course of action, is for the statutory test to be revisited with a view to making it better fit the objectives of the 2002 Directive (as amended). So, for example, a statutory test that was formulated in terms of e.g. annoyance, inconvenience and/or irritation, rather than “substantial damage or substantial distress”, might well have resulted in a different outcome.66. Fourth, and lastly, Mr Cornwell contended that the tribunal had rejected the Commissioner’s guidance without having regard to the fact that the guidance had been approved by the Secretary of State and laid before Parliament and indeed cited an example on all fours with the facts of the present complaint. However, ‘soft law’ comes in many different shapes and flavours, with differing consistencies; some such extra-statutory forms of soft law may be firm, others may be more wobbly. Given the requirements of section 55C(5) and (6), the Commissioner’s guidance is plainly at the ‘harder’ end of the spectrum of such quasi-legislation. Yet in the final analysis it remains guidance, a useful and persuasive tool to interpreting and understanding the legislation. However, such official guidance always remains the servant to, rather than the master of, the primary legislation enacted by Parliament. As Mr Hopkins points out, the Commissioner had in any event agreed in the course of argument in the present case that the approach taken to the meaning of “likely” in the guidance was misplaced. It was the tribunal’s task to apply the statutory test, not the guidance, and the tribunal succinctly explained where and why it differed from the approach taken in the Commissioner’s guidance.67. Mr Cornwell’s fourth and fifth points may also conveniently be taken together. These were the submissions that the tribunal had failed to take into account the direct evidence before it of the distress actually experienced by complainants and, moreover, had substituted its own view of the impact of marketing texts for that of the recipients. In doing so, Mr Cornwell took me to both the spreadsheet categorising and summarising the 286 complaints (annexed to the Commissioner’s further and better particulars, in response to the tribunal’s earlier directions requiring clarification of the contraventions alleged) and a sample of the actual complaints garnered in the course of the Commissioner’s investigation. Mr Cornwell argued that the tribunal had failed to engage properly or at all with this body of evidence, contrary to the requirement to conduct a full merits review. Furthermore, in its conclusions at paragraphs [28] and [29], the tribunal had in effect imposed its own value judgement as to the fortitude that recipients should show when receiving unwanted marketing texts. In effect, as Mr Cornwell put it, the tribunal had concluded that the recipients of these texts had failed to show “sufficient British phlegm”, so ignoring the proper protection afforded to people’s privacy in the measures adopted by the Directives and PECR.68. Mr Hopkins’s answer is that the tribunal plainly did take into account the evidence it received (it referred to the schedule of complaints at paragraph [25]) but expressed caution about the weight to be attached to them. Moreover, the weight to be attached to any particular evidence is quintessentially a question of fact for the fact-finding tribunal (see DBERR v Information Commissioner and O’Brien [2009] EWHC 164 (QB) , [2011] 1 Info LR 1087 per Wynn Williams J at paragraph 32). Furthermore, the complainants’ views (or the views of a sub-set of those complainants) could not be decisive; the tribunal’s task was to review the facts for itself and make its own assessment of the likely consequences of this kind of contravention.[69]I am satisfied that the tribunal took into account the complainants’ evidence. That evidence had been helpfully digested for them in the form of the Commissioner’s schedule to the further and better particulars, as well as being available in its raw form in the multiple lever arch files of individual complaint forms. It is not my role on an appeal limited to error of law to revisit those complaints. Mr Cornwell advised me, and Mr Hopkins did not demur, that oral argument before the tribunal had taken in total about 4 hours. However, I note that the tribunal sat on both 2 and 3 October 2013, indicating that work was done outside the hearing room. I acknowledge that tribunals approach their work diligently. The fact that the tribunal’s decision did not refer to specific individual complaints or indeed to every category of complaint does not mean that the tribunal failed to consider the evidence. As Lord Hope intimated in R (on the application of Jones) v First-tier Tribunal [2013] UKSC 19 , [2013] 2 AC 48 , “it is well established, as an aspect of tribunal law and practice, that judicial restraint should be exercised when the reasons that a tribunal gives for its decision are being examined. The appellate court should not assume too readily that the tribunal misdirected itself just because not every step in its reasoning is fully set out in it” (at paragraph 25).70. Indeed, in a case such as this the tribunal’s task would have become unmanageable if it were required to refer to all the evidence and all the parties’ submissions in its decision. So, for example, the tribunal’s reference to accident claim texts, but not to PPI mis-selling claim texts, in paragraph [28] did not mean that the latter category was overlooked. The tribunal was simply referring to one example (and indeed an example that put the Commissioner’s case at its highest). The tribunal’s job was to review the evidence, apply the relevant law, make its findings and explain them. The Commissioner himself dealt with the issue of whether there was a likelihood of substantial damage or substantial distress in just four paragraphs in the MPN. The tribunal dealt with the matter equally succinctly, explaining why it had come to the contrary conclusion.71. It is undoubtedly the case that the tribunal’s decision in this case was not as detailed as many that emerge from the information rights jurisdiction. True, the tribunal’s decision referred to none of the case law cited by counsel in their submissions, with the sole exception of R (Lord) v Secretary of State for the Home Department . True, the tribunal did not expressly mention that PECR gave effect to the 2002 and 2009 Directives. True, the tribunal did not refer to the concept of privacy, other than in the title of the relevant statutory instrument. But that does not mean that either its analysis or its reasoning was inadequate. It is axiomatic that the question of whether a tribunal’s reasoning satisfies the test of adequacy has to be judged against the background of all the material before the tribunal and the parties’ shared knowledge of the relevant issues.72. In this respect the schedule to the further and better particulars was undoubtedly a key document. It listed all 286 complaints, identified their key characteristics (e.g. type of message content, whether breach of both regulations 22 and 23, etc) and helpfully included the actual text of any free-form comments made by complainants. These were taken from responses to the last substantive question (Q.12) on the ICO’s PECR Complaint form (marketing messages) for members of the public, which asks the complainant “Has the receipt of these messages had any practical impact on you? (e.g. prevented urgent messages from being received, incurred costs etc)”. Complainants are then asked to tick the relevant box for ‘Yes’ or ‘No’ and to provide any details in their own words.73. The schedule shows that some 225 of the 286 complainants did not make any further comment in answer to Q.12. The content and tone of the 61 complainants (i.e. about 1 in 5) who did add such free-form comments varied considerably. Some simply added a short comment to the effect that the texts “irritated the heck out of me” (complaint 91) or were “invasive & offensive when I am busy” (complaint 237). Others made longer entries indicating real personal upset (e.g. “I have severe bi-polar. This is something I am still coming to terms with & find unwanted text messages to a number I really don’t give out very often quite upsetting and disturbing...” (complaint 60). A potential weakness in the complaint form is perhaps that although Q.12 asks complainants about the “practical impact on you”, arguably the focus of the examples given (urgent messages missed, costs incurred) is on damage rather than distress caused. However, the tribunal could only decide the case on the evidence put before it. Taking the schedule as a whole, the tribunal’s finding that the effect of the contravention was likely to be widespread irritation rather than widespread (substantial) distress was plainly one that was open to it.74. It follows that I conclude that the third ground of appeal is also not made out. Conclusion75. I therefore conclude that the Information Commissioner’s appeal must be dismissed. Signed on the original Nicholas Wikeley on 11 June 2014 Judge of the Upper Tribunal

Cited in 2 later judgments