It is important at this juncture to address one of the most common features found in the literature concerned with the defence of privacy. This is the view that privacyisprimarilyconcerned with personal information, and that this is the sole and proper subject of legal protection. We have already seen that Parent advocates such a position, but he is by no means alone.
The works of Raymond Wacks have made a significant contribution to the debate about the propriety and practicalities of privacy protection by legal means.121 In particular, Personal Information, Privacy and the Law sets out 'to obviate the confusion that afflicts the question of "privacy" and obstructs the satisfactory legal protection of the interests with which it is concerned'.122 Thus, while acknowledging that individuals might have a wide range of interests in privacy that can legitimately include, say, the use of contraceptives, termination of pregnancy, or homosexual conduct, he argues that in order to secure viable legal protection of privacy one must start with a workable and relatively narrow concept. Protection, he submits, cannot be achieved under the banner of a wideranging rubric. Rather, Wacks identifies concern with personal information as crucial to privacy interests: '"Personal information" consists of those facts, communications, or opinions which relate to the individual and which it would be reasonable to expect him to regard as intimate or sensitive and therefore to want to withhold or at least to restrict their collection, use or circulation.'123
This is an objective test. It determines the personal nature of information by reference to the reasonableness of the individual's expectations of privacy. Wacks argues that such an assessment is the only viable option, since a subjective assessment of personal information would not only
121 See, for example, 'Privacy in Cyberspace: Personal Information, Free Speech, and the Internet' in P. Birks (ed.), Privacy and Loyalty (Oxford, Clarendon Press, 1997), ch. 4; Privacy and Press Freedom (Blackstone Press, London, 1995); Personal Information, Privacy and the Law ; The Protection of Privacy ; and 'The Poverty of Privacy' (1980) 96 Law Quarterly Review 73.
122 Wacks, Personal Information, Privacy and the Law, p. 1. 123 Ibid., p. 26.
require the protection of 'spurious' privacy interests, but would also require the protection of illegitimate privacy claims, for example, the claim that the fact that X beats her children is personal information. In other words, Wacks' definition sees personal information as something that earns protection by a combination of its quality and the reasonableness of the expectation of the individual claiming protection.124 The reliance on reasonableness is crucial here. In many ways this approach casts a wide net of protection. For example, as Wacks points out, a reasonableness-based assessment of what qualifies for protection allows an individual to claim that even if she has chosen to disclose personal information in one circumstance, it does not mean that she has forfeited protection for all circumstances. You might be happy to tell your doctor that you are infertile but would be very much aggrieved if your employer found out.125
Similarly, Wacks notes that his definition allows an individual to claim protection even if she has no idea that information about herself exists or has been used without authority. This is so because one can determine if the individual might reasonably be expected to view the information as sensitive or intimate 'if she were aware' of the unauthorised activities in question.126 However, the obvious question that this approach needs to address is how the matter of reasonableness is to be determined. In turn, this question can be broken down into a number of related issues. For example, who is to answer the question of reasonableness? Are there any basic preliminaries that mean that the status of'personal information' can be claimed for certain issues as a matter of course? And, is the determination of reasonableness open to influence from the interests of other individuals or from society at large?
Wacks' approach to the practical problem of how to protect personal information is to apply a model for assessing the normative degree of protection that can reasonablybe expected bythe subject. This model depends on the view taken of the sensitivity of the information in question, which he calls 'Information Sensitivity Grading'.127 It works as follows. Different types of personal information are categorised into one of three indices: high, moderate, and low sensitivity. The purpose is to assist in determining the role of the law in regulating the collection and/or the use
124 Ibid., p. 24. 125 Ibid., p. 27. 126 Id. 127 Ibid., pp. 226ff.
of such information. Information concerning how an individual voted at the last general election, how frequently she has sexual intercourse, her divergent sexual habits, her mental health, suicide attempts, misuse of drugs or alcohol and genetic predispositions to illness or congenital handicaps are classified as 'highly sensitive'. In comparison, information relating to an individual's previous address, the fact that she is adopted, her NHS number, her absences from work or her credit rating are examples of information of'moderate sensitivity'. Finally, information of 'low sensitivity' includes sporting activities, membership of clubs, employer details, home address and the fact that an individual wears glasses. Wacks identifies six factors which assist in the categorisation of personal information.128 These are:
• the reasonable expectations of the data subject;
• the recipient of the data;
• the scale of the disclosure;
• the context of the collection, use or disclosure;
• the purpose of collection, use or disclosure.
In this way he provides us with a means of measuring the reasonableness of a claim to protection of personal information through a range of objectively assessed criteria. Even so, the very general question of the reasonableness of the expectations of the data subject remains one of primary importance.
Wacks also accepts that his model is principally normative and not value-free because it classifies information byreference to existing norms and to attitudes towards certain kinds of information. Thus, as he notes, 'medical information accounts for the preponderance of "highly sensitive" data'.129 He also concedes that such a model cannot be exhaustive and will never attract universal approval. This, however, he sees as an advantage rather than a disadvantage: the quality of information can shift as social and political norms themselves change.130
In essence, however, Wacks' classification system is based on a concern with harm. The over-arching question to be asked of information is this: to what extent will unauthorised use or disclosure of the information result in harm to the subject? Thus, the classification of personal
128 Ibid., pp. 2 2 7-9. 129 Ibid., p. 242. 130 Ibid., pp. 226-7.
information becomes a mechanism for risk management and damage limitation. While there is no obvious first-degree objection to this approach, it suggests that information cannot be classed as highly personal if - when objectively assessed - it is thought to cause little or no harm if misused. Such information is therefore unlikely to receive (strong) protection. But, in certain cases, the use or disclosure of particular information that falls outside Wacks' model or that has been classed as 'low sensitivity' might, nonetheless, be harmful for a particular individual, and Wacks' model is unlikely to protect against it. This is not to say that Wacks' adoption of an objective test is in itself wrong. It is simply to assert that a model that is concerned primarily with the avoidance of harm is weakened if it excludes cases where harm is actually likely to result, even if those cases do not fit neatly into the model that has been devised.131 If Wacks' response would be that his list of six factors that determine the category into which any particular form of information should fall will operate in conjunction with the particular interests of the individual - for example, depending on how she views the purposes of disclosure, or how she sees the context of disclosure - then this involves the introduction of a subjective element to his assessment of personal information, and it should be recognised as such.
The consequence of Wacks' system is that the form of protection offered to personal information depends upon the category in which any particular piece of information is placed. He suggests that there is a strong case for holding that information of a highly sensitive nature should, perhaps, not be gathered at all. The implication is that if such information is not gathered, control cannot be lost and therefore harm cannot result from misuse or unauthorised disclosure. Wacks admits that it is difficult to argue that moderately sensitive information should not be collected at all; rather, he suggests that, once collected, it should receive a high degree of protection because 'the potential for harm is of a very high order'.132 Finally, low sensitivity information concerns, in the main, biographical information about an individual, that is, information that is already available in the public domain (for example,
131 Wacks himself admits that his model is concerned with the avoidance and/or reduction of harm: 'The threefold classification used in the above index of "personal information" is based on the extent to which the collection or use of the data holds a potential for serious harm to the subject', ibid., p. 238.
address, telephone number, employment details, etc). The collection of such information is of itself unlikely to cause harm, but the collocation of various pieces of information from different sources can create a category of personal information that is of a higher order of sensitivity. Low order information deserves a degree of protection in this respect.
Wacks' aim is admirable. He offers a model for dealing with personal information that is viable and which provides a relatively sound base for legal supervision. He is, however, fully cognisant of the objections that might be raised and does not purport to offer a definitive account of how personal information should be classified. But, as he himself says, 'the purpose is to demonstrate that personal information is susceptible of this sort of analysis, and that it might offer a more effective means of regulating the collection and use of suchdata'.133
This he does well, and it is not the intention here to criticise that attempt per se. However, there are two objections to the work that can be raised from the perspective of the present analysis. These are, first, the question of third party interests, and second, the restriction of his protection of privacy to personal information.
We have already asked to what extent the determination of reasonableness is open to influence from the interests of other individuals or from society at large. In the first instance, Wacks avoids saying that information in which others have an interest is no longer 'personal' or 'highly sensitive'. His model is designed to apply to all kinds of personal information irrespective of external factors such as third party interests.134 However, he introduces these when discussing the question of whether highly sensitive information should be gathered or used at all. Indeed, he is led to conclude - using the example of persons afflicted by AIDS -that legitimate third parties' interests should be furthered if possible. He then argues that to say that highly sensitive personal information should not be collected at all is no answer to the question of its appropriate legal protection. His solution in the context of AIDS is anonymisation. In that way, he submits, the risk of harm to the individual is minimised but, at the same time, the public interest in statistical data on the spread of HIV and AIDS can be furthered. Unfortunately, such a solution cannot be applied across the whole health care spectrum. In particular, it is not open in the context of genetic information, where we are concerned with
the interests of family members as well as those of the individuals who have been tested for genetic disease; we face the very difficult problem of determining to whom genetic information relates. This is explored further in chapter 5.
Wacks' analysis proceeds on the assumption that personal information concerns one person, who, as a consequence, should control the information in question. But as we will see in chapter 3, the position is considerably more complex in the case of genetic information when the essential nature of the information shifts from the personal to the familial. Wacks' model does not give us a means of assessing such circumstances. This can be seen, inter alia, from his views on the use of medical data. He posits that one approach to the problem of the legitimacy of using highly sensitive medical information is to argue that its use is in the patient's best interests.135 He rejects this, however, as paternalistic and prefers instead the option of seeking the patient's consent. This is viable in the context of the single doctor/patient relationship, but individual patient consent provides us once again with a problem in the context of familial genetic information. Even if patient A refuses to allow access to genetic information, how are the interests of A's relatives, B and C, to be assessed when they might want or need to know the information? Wacks' model lacks the sophistication to address such problems.
The second objection to be raised to Wacks' thesis relates to his decision to restrict privacy protection to personal information. We have already considered his reasons for doing so: such a construct encompasses most concerns that are expressed in privacy terms, and Wacks suggests that such a concept is more amenable to effective legal protection. His aim is to strike 'the best deal' for a plurality of privacy interests. In the main, he puts forward a convincing and important argument, one that has a sound basis and which provides much of practical utility for those considering the problems of legal protection of privacy. However, it is a primary aim of this work to show that privacy concerns are not restricted simply to concerns about information and further that to fail to recognise this leaves many important non-informational interests unprotected. As an illustration of this consider two examples that cause problems for Wacks' model: the permanently insensate patient136
136 The permanently insensate patient is, for example, one in irreversible coma or permanent vegetative state (PVS).
and the individual who does not want to know about her own genetic constitution.
In the case of the irretrievably insensate patient, how is such a person harmed by the taking of information about her? She is not conscious, she will never regain consciousness and the information that is obtained is completely beyond her control. Wacks' model would lead us to classify all information about such a person as of low or no sensitivity because the person cannot be injured by the misuse of the information. Yet intuitively we would still say that the patient's privacy had been invaded if, for example, newspaper reporters burst into her hospital room and took photographs and recorded details of her condition.137 Surely, what is invaded here is not privacy in information, but rather privacy in the self -a sphere of physical privateness that others should not invade without authority and good reason. Moreover, such a sphere exists irrespective of whether the subject can control it or is conscious of its existence. This is so because that sphere is linked not just to information and fear of harm, but to individuality, dignity and respect for persons. This respect can be measured purely against the conduct of the aggressor, and can accordingly be found wanting, even when the aggressee is unaware of any intrusion into her private sphere. A harm-based analysis such as that of Wacks requires a degree of likelihood of harm before privacy interests are considered deserving of protection - no likely harm, no protection. Moreover, Wacks' step-like approach by way of the sensitivity of information does not help us here, because the person to whom the information relates is unaware, and unlikely ever to become aware, of any intrusion.138 Thus, even on an objective analysis, there would be no likelihood of harm and therefore no reason to protect the information.
The individual who does not want information about her own genetic make-up is also unlikely to be protected by Wacks' analysis. His entire assessment of sensitivity is concerned with harm arising from unauthorised communication of personal information to third parties. However, the individual who does not want to know about familial genetic information is concerned that information about herself should not be communicated to herself. Moreover, and as will be argued further in chapter 3, it is not clear that the interest that an individual has in not knowing information is to be properly regarded as an informational
137 Note the parallels here with Kaye v. Robertson  FSR 62, discussed in ch. 1.
138 Wacks, Personal Information, Privacy and the Law, p. 27.
privacy interest. Were it to be so, the nature of privacy would, once again, be reduced to a control-based concept. The individual's privacy argument would then be premised on a claim to control information about herself and, for the reasons articulated above, that is an unsatisfactory view of privacy. Furthermore, a claim to control personal information presupposes that the subject of the information knows the information to exist. The most likely scenario in this context is, however, one in which the subject has no idea about the information in question, for it will have been generated outside her immediate ambience, for example through the testing of relatives; and a person is powerless to exercise control over something about which she is ignorant. Rather, if privacy is seen as a state or condition - which it is submitted is the preferable view - then it becomes a sphere of separateness that deserves recognition as being inviolable, and not merely an issue of control of information. That state of inviolability, if respected, would protect an interest in remaining in ignorance of one's own genetic constitution. This view is akin to that of a spatial privacy interest, something which Wacks would not accept as part of his model.
The above comments should be taken not so much as criticisms of Wacks, but as defences of the approach adopted in this work, which is to argue a view of privacy that encompasses both informational and spatial privacy interests. The utility of this lies in the fact that it widens and, at the same time, crystallises the range of interests that are legitimately subsumed under the privacy rubric. Less ambitiously, it highlights the strong arguments that can be made against narrower constructs, even if in the short term it must be conceded that protection along the lines suggested by Wacks is a necessary and important first step. But, even if that is the case, we should not stop at that point. We should not rest on our laurels simply because we have found a way to protect some privacy interests. This work posits that the limits of the law can be pushed further forward in the pursuit of adequate protection for legitimate privacy interests. It paves the way for future and further steps along the privacy path. For these reasons, Wacks' narrow construction of privacy is rejected. And yet, the aim of this work and that of Wacks can be reconciled in the following paradoxical fashion. Wacks seeks to extract a narrow view of privacy from the literature in order to broaden the range of protection that can be afforded to privacy by law. This work examines a narrow area of privacy concern, namely genetic privacy, in order to broaden the privacy debate. The precise role for the law that emerges from this analysis will be examined more fully in later chapters.
Was this article helpful?