Annotated Bibliography

GROUP PRIVACY AND GOVERNMENT SURVEILLANCE OF RELIGIOUS SERVICES*

Privacy is not typically seen as a 'religious' issue, but faith groups in Canada are going to have to address some difficult questions relating to privacy in the near future, if they are not embroiled in them already. A par- ticularly pressing area of concem is government surveillance of religious services, and I think here especially of Canada's Islamic community. If CSIS' were to send undercover agents to attend services at mosques and monitor sermons given by Canadian Imams, in the hopes of spotting nascent terrorist sympathies or recruiting tactics, could this constitute a loss of privacy on the part of the community of faithful gathered at the mosque? Leave aside for a moment the question of whether, if a loss, it would be justified and hence not a violation of privacy. Is govemment sur- veillance of communal religious services even a privacy issue? After all, such services are typically seen as public events, which in the eyes of many would disquahfy surveillance of them from entailing a loss of pri- vacy on the part of those in attendance, whatever other moral or legal issues it might raise. Indeed, the very notion of 'group privacy' may appear an oxymoron. Nonetheless, such surveillance is a privacy issue, and in what follows I hope to show this and to map out some of the con- ceptual contours of privacy in this context.

In doing so, we must first proceed through a brief analysis of the con- cept of privacy itself. Some comments will then be made on privacy-loss and violation, followed by a discussion of the distinctions between insti- tutional, group, and individual privacy and the relevance of these distinc- tions in the present context. Special attention will be given to whether a real distinction can be maintained between the latter two types of privacy. I will argue that one can properly speak of a group possessing and losing privacy, and having its privacy violated. Finally the specific application of group privacy to communal religious services in a mosque will be

"Group Privacy and Govemment Surveillance of Religious Services" by Travis Dumsday, The Monist, vol. 91, no. 1, pp. 170-186. Copyright © 2008, THE MONIST, Peru, Illinois 61354.

GROUP PRIVACY AND GOVERNMENT SURVEILLANCE 171

addressed, and it will be seen that surveillance of such a service could constitute a loss of privacy both on the part of the group and the individ- uals constituting it, making such surveillance a privacy issue. My prima- ry concem throughout is conceptual rather than normative, though towards the end I will comment on the ethical status of such surveillance.^

/. The Concept of Privacy

To start off, I must provide a definition of privacy, and the one used here is essentially informational. It seems to me that accoimts which make privacy a matter ofthe possession of information are the most intuitively satisfying of the various options on the market. And indeed, there are a number of competing theories of what privacy really is.^ For reasons of space I cannot provide here a defence for my view of privacy as informa- tional, but suffice it to say that a convincing case has been made for this by Davis, among others."* I will therefore work with the following defini- tion of privacy, one similar, though not identical, to that employed by him:

Person, group of persons, or institution X has privacy with regard to fact P and with respect to person Y if and only if P is not known by Y and is, in some way, sensitive information (i.e., information which, if generally revealed, would likely result in some damage or harm to X, whether physical, emotional, financial, or other).

Regarding the 'sensitive information' clarification, it is not necessary that Y or others should actively employ the revealed information to harm X in order for that revelation to result in harm on the part of X; it could be that the simple fact of Y becoming acquainted with fact P results in emotional suffering for X. And no actual harm or damage is even required; just the likelihood of it. The 'generally' is inserted to cover what Parent notes con- cerning this type of information, namely that people "may not be con- cerned that a few close friends, relatives, or professional associates know these facts, but they would be very much concerned if the information passed beyond this limited circle.''^ Of course, information is that much more private if its revelation even just to one's inner circle would likely cause harm. I have used 'sensitive' rather than 'personal' to denominate this type of information, as 'personal' would seem to disqualify at the out- set groups and institutions from privacy possession

172 TRAVIS DUMSDAY

One final point: the content of the information need not be something most people would regard as somehow inherently 'private' in order for it to meet thé above Criteria of sensitivity. Consider a person's height; for most people, their exact height is not a piece of sensitive information, but for some it may be, and as a result those people can possess privacy with respect to that information even if the rest of society does not. "Consider, for example, the man who is ultra-sensitive about being short and who tries desperately to conceal his actual height from even his closest friends. It makes perfect sense to say that this person's height is for him personal information. My discovery by ingenious snooping of just how short he is could plausibly be called an invasion of his privacy. "«

2. Privacy-loss and Violation

With these conceptual tools in hand, we can proceed to our main topic: whether govemment surveillance of communal religious services, in this case services at a mosque, is a privacy issue. And that is essential- ly a question of whether such surveillance can result in a loss of privacy on the part of the group and/or its individual members. So it would first be a good idea to provide a bit more detail on privacy-loss, as well as its distinction from violation. Referring to our technical definition, we can say that X undergoes a loss of privacy with respect to sensitive fact P and with regard to person Y if and only if Y comes to know P. What of viola- tion? A loss of privacy does not, in and of itself, have normative implica- tions. Privacy can be willingly relinquished, and usually such a loss will not be thought to involve a violation of one's privacy. It is only when this loss occurs in an immoral fashion that a violation takes place. It would be difficult to give a precise definition of 'immoral fashion', given the almost infinite variety of ways in which a person's privacy might be violated; suffice it to say that 'immoral fashion' must be interpreted broadly, such that it encompasses more than just the means by which the sensitive infor- mation is obtained. Information can be obtained in a legitimate manner yet still involve a violation. If Y is walking down the street and comes across X's diary, it is not unethical to pick up the book and examine it in order to see what sort of book it is and who it might belong to. The diary is thus obtained ethically. But once Y realizes it is a diary, it is certainly immoral for him to walk home and read through it. X has not given con-

GROUP PRIVACY AND GOVERNMENT SURVEILLANCE 173

sent, and the diary likely contains sensitive information, so to read it con- stitutes a violation of X's right to privacy even if it was obtained licitly. Moreover, there are times when information is willingly relinquished but in which privacy is still violated. If X goes to confession and confesses to someone he believes to be a priest, but who in fact is an imposter. Y, who gets a kick out of hearing people's confessions, a gross violation of priva- cy has obviously taken place.

3. Institutional, Group, and Individual Privacy

We now turn to the distinctions between institutional, group, and individual privacy. My main concern is with surveillance of the actual communal service at the mosque. Other forms of surveillance will not be under consideration, as it seems to me that it is this type which most involves the religious community itself in privacy-loss. And in order to be clear on the question of what exactly is occurring in this type of surveil- lance, it will be useful to parse out these three varieties of privacy.

Consider, for example, a case in which the govemment has decided that a certain Imam is worthy of special attention as a potential security threat. As a result they bug his place of residence and car. Is this a priva- cy issue? Most certainly, though in this case it is the Imam's individual privacy that is at issue, not the privacy of the Muslim community gathered at the mosque. What if the government bugs his office at the mosque itself? Here we land in murkier waters. This involves a loss of privacy on the part of the Imam, but is more involved? One might think that the bug- ging of the mosque itself, even if only in the private office of the Imam, implicates the mosque as an institution. As a result, the privacy of the institution comes into play, and not just the individual privacy of the Imam. This was not so with the bugging of his home or vehicle. Yet this might not always be the case; if the Imam's office is bugged because of a perceived threat unrelated to his institutional role, for instance if he is sus- pected of being involved with a bank robbery, then institutional privacy per se is not at issue. The bugging of the office in that case would involve the institution in only an accidental sense; it would be merely one addi- tional means of getting at the Imam. However, group privacy would not come into play in either instance, indicating one context in which the dis- tinction between institutional and group privacy is evident.

174 TRAVIS DUMSDAY

I do not wish to delve into the topic ofthe distinction between group and institutional privacy in any detail, as my principal concem is with the distinction between the former and individual privacy. Yet it is worth not- ing that institutional privacy is dependent on, though not entirely reducible to, that of the group. The existence of the institution is depen- dent on its members, who founded and maintain it, though some distinc- tion between the two can still be held. The institution is best thought of as an instrument of the group, as a means of carrying out the aims of group members in a concrete manner. An instrument is in a strong sense depen- dent for its operations on the individual or group wielding it, yet it remains separate from them. Thus the distinction between group and institutional privacy can be maintained, and it is not the case that all instances of insti- tutional privacy-loss will constitute a privacy-loss on the part of the group. The bugging ofthe Imam's office is one example of this. Nor is it the case that all occurrences of group privacy-loss will constitute institu- tional loss. If the members ofthat Muslim community leave the mosque after a worship service and travel to a private home for lunch, and that home is under surveillance for reasons unconnected with the mosque, the group will come under surveillance and hence experience a loss of priva- cy, without the institution to which they belong and from which they have just come being directly implicated.' Group privacy-loss is thus neither a necessary nor sufficient condition for institutional privacy-loss, and the same holds vice versa. They must therefore be distinct.^

This brings us to group and individual privacy. Both Westin and Davis advocate the inclusion of groups as bodies capable of possessing and losing privacy. I too have included it in the definition of privacy, but so far have left it as an unargued postulate. As a preliminary to its defence I should first state what is meant by 'group' here. The literature in the phi- losophy of social science is replete with competing definitions of groups and disputes about whether or not they even exist. But I believe it is intu- itively plausible that groups do have some sort of existence, however that is to be cashed out. For present purposes I believe the existence of groups can be taken for granted, and May's definition will suffice: "when a col- lection of persons displays either the capacity for joint action or common interest, then that collection of persons should be regarded as a group."'

Is the application of privacy to groups in need of defence? Perhaps, though before proceeding to look at such defences the intuitive appeal of

GROUP PRIVACY AND GOVERNMENT SURVEILLANCE 175

extending privacy to groups ought to be noted. Consider a secret society like the Freemasons. Imagine that person Y disguises himself and sneaks into one of their meetings, observing some of their most guarded rituals. I believe it is prima facie reasonable to speak of the Masons as a group undergoing a loss of privacy, and not just each individual member. "̂

One way to challenge this view would be to claim that the so-called privacy of the group is nothing more than the individual privacy of its members, taken together. Thus, each individual Freemason possesses pri- vacy with respect to his own actions in and presence at the secret meeting, and whatever privacy the group might be thought to enjoy is really just the accumulated privacies possessed by each separate member. On this view, group privacy is reducible to individual privacy.

Davis addresses this reductionist argument. He claims that a group can experience a loss of privacy even when each of its members retains his own anonymity, which is often thought to be a guarantee of privacy: "Suppose that there is a secret society that has rituals, information about which the society wishes to keep from the public eye. If the rituals of the group were reported in the newspaper, but no names of members were reported, the group could claim that it had suffered a loss of its privacy, but no member could so claim a loss."" This seems to show that a group can experience loss independently of any loss of privacy on the part of its individual members, so group privacy is not reducible to individual privacy.

In order to avoid problems that arise from the notion of anonymity, Davis's example can best be interpreted such that what the newspaper reveals is not the details of any particular performance of the secret ritu- als, which would implicate particular individuals in attendance, but rather the details of the rituals in general, as they might be performed at any gathering of the group.'2 Perhaps the newspaper obtained one of the soci- ety's manuals of ritual. In this case, no particular individual is implicated in the newspaper's revelations, even anonymously. What is revealed is rather a set of facts about the group as a whole, and we have an issue of group rather than individual privacy. In fact, we can imagine cases in which none of the members of the group could possibly be implicated in the newspaper's revelations, thus disallowing any sort of reduction to individual privacy-loss. Consider the case in which a brand new secret society has just been founded, and the eager new members of the group, none of whom have yet had the chance to participate in one of its rituals.

176 TRAVIS DUMSDAY

are studying their copies of the manual in preparation for their first offi- cial gathering. On his way home fi-om the study session member X acci- dentally leaves his copy of the manual on the bus, where it is seen and picked up by reporter Y, who, realizing the hot story, quickly does a write- up and publishes extensive excerpts ofthe manual in the morning edition. In this instance, no member of the group has ever had a hand in the con- ducting of one ofthe rituals, so no fact about any ofthe individual mem- bers ofthe group has been published. What has been revealed is informa- tion about the group itself, not about any one of its members. In this inter- pretation of Davis's example we have an instance of group privacy irre- ducible to individual privacy. Thus a real distinction between the two must be maintained.

A potential criticism would be that at the very least the privacy ofthe individual member or members who had a hand in writing the new man- ual would be compromised by the newspaper's action; for though they remain anonymous with respect to it, anonymity is not sufficient to guar- antee privacy with respect to all types of information. But the example can be altered to eliminate this concem. Assume the manual was written by a non-group member, perhaps an outside consultant who was brought in specially for the purpose, or a member who quit the group after the writ- ing ofthe manual but before its revelation by the newspaper. Posit that for these people, the manual and their authorship of it is not in any way sen- sitive information; they would not be harmed by its revelation, and as it happens could not care less one way or the other. In such cases whatever privacy is lost must be that ofthe group and not of any individual.

One might yet maintain that group privacy is nevertheless hopeless- ly bound up with individual privacy. If X is a member of a secret society, then whatever happens to the group, even just as a group, implicates X by virtue of his membership in it, with the result that one can interpret priva- cy-loss in this context as still being individual. The idea is that X's iden- tity extends to include the group, and facts about the group will thereby be facts about X. Consider what Gavison has to say regarding privacy and information extension:

Problems ofthe relationship between an individual and pieces of information exist on another level as well. Is information about X's wife, car, house, par- ents, or dog information about X? Clearly, this is information about the other people, animals, or things involved, but can X claim that disclosure of such

GROUP PRIVACY AND GOVERNMENT SURVEILLANCE 177

information is a loss of his privacy? Such claims have often been made. Their plausibility in at least some of the cases suggests that people's notions of themselves may extend beyond their physical limits.'3

I am not convinced that this sort of extension can work, and suspect that a distinction needs to be introduced. The fact that I used to own a rabbit is definitely a piece of information about me. Is the fact that Fuzz was a Netherland dwarf rabbit infonnation about me? No, unless I am in some way contained or implicated in the proposition expressing the informa- tion. "Fuzz was a Netherland dwarf rabbit" is information about Fuzz. "I had a rabbit named Fuzz who was a Netherldand d w a r f is information about me. The difference between the first and second statement is clear.

However, even if the point holds, and information about the group is automatically information about X due to X's membership in it, I do not see how this is a strike against the notion of group privacy. Quite the opposite, in fact, for here the privacy-loss on the part of the group causes the privacy-loss on the part of its member. If X were not a member of the Freemasons, then he would in no way be implicated in the group's priva- cy-loss should some secret about it be revealed. It is the group's privacy- loss, and X's membership in the group, that causes X's privacy-loss. So if this notion of information extension is correct, the result is that every loss of group privacy will also be a loss of individual privacy, contradicting some of the arguments made above. But it also has the result that group privacy is just as real as individual privacy and cannot be reduced to the latter, since it causes the latter in such cases. For reduction to be possible, the dependence would have to flow in the opposite direction.

So groups can possess privacy, and as the thought experiments about the secret society show, they can also lose it. Since they may lose it in an immoral fashion, a group may also be the subject of a violation of priva- cy. Note that here I am avoiding using the language of rights; I have not said that a group may have its moral right to privacy violated, but that it may have its privacy violated. Why the difference? According to my ear- lier definition of privacy violation, no rights need to be invoked. There may be an advantage to employing rights discourse in the context of pri- vacy, and certainly it is often done. But I do not believe it is essential, even when speaking normatively. Further, doing so here would be problematic, as there are some awkward conceptual and normative problems that arise with the attempt to assign rights to a group.'"^ However, it does seem plau-

178 TRAVIS DUMSDAY

sible to think that there might arise situations in which it is immoral to weaken, damage, or break up a group by releasing sensitive information about it. And that is all I require in order to say that a group may have its privacy violated.

To recap: Real distinctions exist between institutional, group, and individual privacy. There are instances in which a group's privacy is vio- lated although no individual's privacy is violated, and vice versa. With these distinctions in hand, we can now turn to the special case of commu- nal religious services and attempt to discern what sort of privacy and pri- vacy-loss might be involved in govemment surveillance of them, if any.

4. Privacy and Communal Religious Services

Can govemment surveillance of communal religious gatherings even be considered in this light? After all, are they not public events? If so, then presumably for the govemment to monitor their proceedings could not be a violation of privacy. It might perhaps be imprudent or offensive in var- ious ways, but not a violation of privacy. I think this is prima facie a plau- sible argument, but one which ultimately fails. To show this, several ques- tions must be addressed. One is whether any privacy violation can take place in a public context. If this is possible, then it is possible of a public religious gathering. Another is whether, and perhaps to what extent, some religious gatherings, in this case services at a mosque, are tmly public.

To the first question: It is clear that violations of privacy can occur in a public setting. If Mrs. Jones stands up at a town-hall meeting and tells of how her neighbour's husband is having an affair, it is plausible to think that privacy violation has just occurred. So if private information is revealed in public, the fact that it is in a public setting does nothing to mit- igate the violation; quite the opposite, in fact. So even if a communal reli- gious gathering is in the fullest sense a public event, that does not guar- antee that no privacy-losses or violations can occur in this context.

In addition, privacy-loss can occur via govemment surveillance in a public setting. In the middle ofthe public square, a man sets up a podium and, through a loudspeaker, declares the moral necessity of buming down the local police station, as the police are the instruments ofthe corrupt ml- ing classes and it is high time the proletariat took a decisive stand against the bourgeoisie. He announces his intention to commit the act and solicits volunteers. An undercover officer happens to be present and takes note of

GROUP PRIVACY AND GOVERNMENT SURVEILLANCE 179

all this. Now, has a loss of privacy occurred? On the part ofthe man at the podium it certainly has; he has relinquished his privacy to the crowd with regard to the sensitive fact that he intends to bum down the local police station. No violation has occurred, because the information was relin- quished willingly. So individual privacy-loss has taken place. What about group privacy-loss? The crowd in the square was not assembled there for any joint purpose; they did not know ofthe fellow's upcoming declaration and were not there specifically to hear it. The crowd forms a group only in a tenuous sense. So it is doubtful that group privacy-loss could occur here in any but a correspondingly weak sense, unless perhaps this random assembly happens by some chance to share the opinions of the speaker and begins cheering and signing up for the project. If that takes place and is noted by the officer, group privacy-loss may occur, as the new-found com- mon purpose adds another relation between those present, perhaps suffi- cient to constitute a group in a stronger sense;'^ and individual privacy-loss will be undergone by any particular crowd member in relation to the officer and with regard to the fact of his support, if the officer observes him.

Applying these observations to a communal religious service, we see that, even if it is a fully public-event, privacy-loss can occur in it, both individual and group. If an undercover officer attends a service at a mosque for the purpose of monitoring the Imam's sermon and the group's reactions to it, both types of privacy-loss may occur; this is especially the case as the group of people gathered at the service must surely be seen as a group in a very strong sense. Yet privacy will only be lost if sensitive information is revealed. Whatever the Imam says to the group in atten- dance, he loses his individual privacy with respect to what he reveals thereby and in regard to the group, but only if what he says constitutes sensitive infonnation. If there is a government agent infiltrating the group, the Imam also loses his privacy with regard to the agent. And if the group's reaction to the Imam's revelation of sensitive information also conveys such information, the group will have lost its privacy with respect to that information and in regard to the officer. If the officer happens to take note of particular members of the group and their reactions, their individual privacy will have been lost in that regard as well. Yet if no sen- sitive information is revealed, no privacy-loss occurs.

The upshot of all this is that the mere presence of a govemment agent at the service and his monitoring of what goes on there will only involve

180 TRAVIS DUMSDAY

those present in privacy-loss if sensitive information is revealed. If the Imam says nothing the govemment finds objectionable, the agent's obser- vations will not constitute a loss of privacy on the part of the Imam, the group, or any individual members of the group with regard to the govem- ment. So if a religious service is a public event, the mere presence of gov- emment monitoring is not enough to guarantee that privacy-loss occurs. Rather, whether or not it does is contingent on whether sensitive informa- tion is revealed. If such information is revealed, loss of privacy on the part of the individual, the group, or both will occur. And note that what con- stitutes sensitive infonnation will vary according to context. If the Imam condemns suicide bombers, the information conveyed thereby is not sen- sitive with regard to a Canadian govemment agent. If an Imam in an Afghan mosque during the reign of the Taliban delivered the very same message, it would likely constitute sensitive information with regard to an Afghan govemment agent. Thus, while surveillance of a public event is not enough to guarantee privacy-loss on the part of those present, it may result in privacy-loss to individuals, groups, or both.

So even if a mosque service is a public event, it is possible for pri- vacy-loss to occur via govemment surveillance. But is the typical service at a mosque a public event?'« If not, the conclusion that govemment sur- veillance can result in privacy-loss on the part of the group in attendance will be strengthened. The answer may depend on what counts as 'public,' and here we approach another tricky conceptual problem. To be brief, I think that sufficient conditions for a gathering to be public would be if it were held on public property and advertised as open to anyone with no explicit conditions of entry. A public town-hall meeting, for example, or an organized and free gathering in a public park. But these are obviously not necessary conditions, for a public gathering can be held on private property. A necessary condition is more difficult to come up with, but I think a plausible candidate would be that a gathering is public only if it is open to anyone. More detailed specification is no doubt required here, but what I mean is something like a gathering in which no one is excluded on some specific grounds, whether explicit or implicit, such as being a woman, or of a certain race or political affiliation. Any meeting in which such exclusions are made cannot properly be termed 'public'.

So is a worship service at a mosque a public event? Well, certainly no one is excluded on grounds of race or gender. Non-Muslims are wel-

GROUP PRIVACY AND GOVERNMENT SURVEILLANCE 181

come as well as adherents. Yet, although the services may be norhinally open to anyone, it can be argued that there is an implicit understanding present whereby those in attendance at a worship service are there for friendly or at least neutral reasons (curiosity, for instance). If someone attends the service for potentially hostile reasons, this understanding is breached. It is not unreasonable to think that someone would be excluded if it were known that he was there on behalf of CSIS to collect informa- tion for the govemment. Even if there is nothing whatever to hide, it might be seen as improper or as a breach of trust. One could say then that the gathering is restricted on grounds of employment, or perhaps motivation of the attendee. Thus the service is not a public gathering in the same sense as the town-hall meeting would be, in which a CSIS agent presum- ably could not be excluded even if his presence were known, indeed even if he were there on behalf of CSIS, however uncomfortable it might make the other members of the public and the municipal officials.

So a mosque service is not a public event, or at least not fijlly public, if indeed it makes sense to speak of degrees of publicity. As such the claim that privacy violation can occur in this context is strengthened. However, this might be taken too far. It might be argued that such an unwelcome individual as a government agent there for the purposes of surveillance would violate the privacy of those in attendance simply by virtue of his presence. Consider an analogy: if person Y sneaks into a Freemason meet- ing and pretends to be a Mason, he would be thought to be violating that group's privacy. If Y in bad faith and under misleading pretenses attends services at a mosque, this would be seen as a similar violation. This is the case even though the event is nowhere near as private as the Freemason gathering; it is still private to some extent, by virtue ofthe implicit exclu- sion of certain people, namely those of bad faith or inappropriate motives. This is an exclusion which would not apply in the context of more or fully public gatherings, such as the town-hall meeting.

Against this I would reiterate the fact, that in order for privacy to be lost, sensitive information must become known. The infiltrator ofthe Masons is almost bound to gather such information, because any facts revealed about them makes them less secret and damages their very character as a secret society. However, if the infiltrator attends the meeting but is forced to wear a blindfold and earplugs throughout (perhaps before the start of the proceedings someone questions his Masonic credentials), no privacy-

182 TRAVIS DUMSDAY

loss will occur; his presence alone cannot bring about such a loss, and without a loss there is no violation. These observations apply all the more forcefully in the case of a mosque, which hardly classifies as a secret soci- ety or one in which ahnost any information revealed will be sensitive. The mere presence of a govemment agent is not enough to cause a loss of privacy.

Yet what of the agent who, with no proper justification (see below), spies on a worship service and hears nothing? No privacy-loss and hence no violation occurs. But surely some wrong has been committed. I would say that what has taken place is an attempted violation of privacy. This, too, is morally wrong, but ought not to be confused with privacy violation strictly considered.

5. Normative Implications

So far the discussion has been conducted largely apart from norma- tive considerations, and it should not be thought that, simply because gov- emment surveillance of a mosque's services does not necessarily result in a loss of privacy (and hence not necessarily a violation), it therefore rais- es no ethical problems. Far from it. To begin, such surveillance may involve deception. If the agent actively pretends to be a Muslim when he is not, for instance (perhaps to gain the confidence of the Imam and access to more information), this will be deceptive. Even if the govemment engages an actual member of the congregation to report back to it, it could still be argued that a breach of trust takes place, a breach between that member and the rest of the group. Such deception and breaches of trust might be justified in some circumstances, but they do re^M/re justification, and indeed fairly strong justification, if they are to be carried out without acting unethically. Especially strong justification is required in this con- text because services of religious worship are worthy of a special respect. Services at a mosque are being conducted with a specific set of purposes in mind: the worship of God, the edification of those in attendance, the building up of a community, etc. These goals are not to be trifled with lightly. And though it may be a controversial point, I would argue that these purposes make a religious service more worthy of particular regard in this context than, for instance, a Masonic meeting, or for that matter a political or charitable association gathering. Religious faith tends to be much more closely and resolutely bound up with a person's core identity.

GROUP PRIVACY AND GOVERNMENT SURVEILLANCE 183

such that interference with religious worship is a greater affront to indi- vidual dignity than obstruction of other communal activities. If such sur- veillance were to be instituted, and the group were to be alerted to it some- how, worship would inevitably be disrupted for some. How could it carry on as normal when the members realize, or strongly suspect, that one among their number is a govemment agent there to monitor their words and actions? Even if possible, it would hardly be easy. Such surveillance there- fore, even when it does not result in a loss of privacy, may impinge on an activity vital to the identity ofthe group and the individuals comprising it, and may likewise constitute an attempt at privacy violation. If it is to be conducted, therefore, it should only be for the strongest of reasons.

What would constitute justification? Only a genuinely plausible threat to public security. The Canadian Muslim community is over- whekningly law-abiding, and I believe it is safe to say that the large majority of Canadian Muslims are not susceptible to extremist views and terrorist recmiting. However, recent events in which a group of Muslim youth were shown to have been active in planning a terrorist attack on Canadian soil highlight problems faced by that community in keeping out extremist influence (influence which may be organized and funded by overseas extremist groups). Such influence poses a threat to the public, and a sufficiently grave threat to warrant some govemment surveillance. But what sort of surveillance? It is doubtful that justification could be found for routine surveillance of mosques in general. Some are overseen by Imams who have taken vocal public stances against extremism; unless it is believed (on good evidence) that this is some sort of façade, it would seem excessive for the govemment to have agents monitor services and other activities at such mosques. In these cases, if surveillance is to be conducted, it must be in response to a specific, strong indication that something criminal is occurring. Perhaps a tip from members ofthe com- munity itself, or others who have a good level of awareness of what is happening at the mosque. But blanket surveillance, absent such an indi- cation, is morally wrong. Moreover, it could contribute to the very radi- calization that the surveillance is seeking to combat, for if the communi- ty were to leam of this widespread govemment interference there would be an increased danger that the small number prone to radicalization would be pushed over the edge by this perceived affront to Islam and antagonism

184 TRAVIS DUMSDAY

toward the community. It might also create enough general animosity among Canadian Muslims that the community would be less likely to cooperate constructively with the authorities in working to stamp out extremism. Covert surveillance is hardly renowned for its abilities to build trust and foster good working relationships.

However, what of those small number of mosques which are known to have a history of problems in this area? Here it is morally acceptable (though distasteful and regrettable) to engage in routine surveillance. This must be done with every effort to avoid interference with the worship and ongoing life of the community gathered at the mosque, and this requires maintaining absolute secrecy. As far as possible, the congregation should not know that surveillance is taking place. But if it becomes known, the authorities must work to emphasize to the Muslim community as a whole that only those communities with known problems were being targeted, in order to minimize the fallout.

I would like to stress here that this provision for monitoring groups with a history of proneness to extremism is hardly limited to the Muslim community (and again, for the most part, such a proneness is decidedly absent among Canadian Muslims). The fact is that any religious, political, or other group with histories of problems ought to be subject to govem- ment attention. A comparative example would be those cults which have a record of violent behaviour. If a new branch of the Solar Temple or Heaven's Gate cult were to land on Canadian soil, it would only be pru- dent for law-enforcement officials to keep close tabs on their activities. Though even in these instances, care must be taken not to overstep the rea- sonable boundaries of ethics and law.

6. Conclusion

Over the course of the preceding pages we have seen that privacy, properly defined in terms of sensitive infonnation, can be possessed by institutions, groups, and individuals. It can likewise be lost and violated. Govemment surveillance of communal religious services (whether con- ceived as fully public or only partially public events), can result in priva- cy-loss and violation both for individuals and the group, and even when surveillance does not have these results it can be questioned ethically and legally. This makes a strong justification requisite for such activity, and

GROUP PRIVACY AND GOVERNMENT SURVEILLANCE 185

safeguards will be needed to ensure that the justification in any particular case is adequate.

Travis Dumsday University of Calgary Alberta, Canada

NOTES

* A version of this paper was presented at the philosophy graduate student collo- quium of the University of Calgary in the fall of 2006, and I would like to thank those in attendance for their comments. I would also like to express my sincere thanks to Steven Davis and Brian Orend for their feedback on earlier drafts, and to David Matheson for helpful discussions on privacy.

1. The Canadian Security Intelligence Service, akin to the American FBI, although they also engage in foreign intelligence gathering.

2. I should note at the outset that I am using the Canadian Islamic community as my example here and throughout the paper principally because I believe they are the most likely community at present to be faced with the prospect of such surveillance. This is not to suggest that other communities might not be subject to the same attention either now or in the future, nor am I implying that the Islamic community is more deserving of govem- ment attention than some other. It seems that governments and the press have a short mem- ory; not so long ago in the U.S. it was domestic militia groups that were the chief object of concem, with cults perhaps running a close second. In the 1980s the Canadian govem- ment was focused on Sikh terrorists in the wake of the Air India bombing. This is not to say that officials were unwarranted in targeting these various groups for attention, or that they should not be devoting resources to combating Islamist terror cells. I am simply point- ing to the fact that Islamist terrorism may tum out to be a comparatively short-lived phe- nomenon, at least in the West. Who knows what the principal focus of domestic security agencies will be fifteen years from now?

3. For the view that privacy is a matter of being left alone, see Samuel Warren and Louis Brandeis, "The Right to Privacy," The Harvard Law Review, 4 (1890), 193-220; for the view that it is a matter of control over information rather than possession of it, see Alan F. Westin, Privacy and Freedom (New York: Atheneum Press, 1967); for a view of priva- cy as limitation of access to an individual, refer to Ruth Gavison, "Privacy and the Limits of the Law," The Yale Law Journal, 89 (1980), 421-71.

4. See Steven Davis, "Privacy, Rights, and Moral Value," University of Ottawa Law and Technology Journal, 3 (2006), 109-31; W. A. Parent, "Privacy, Morality, and the Law," Philosophy and Public Affairs, 12 (1983), 269-88; and David Matheson, "Unknowableness and Informational Privacy," Journal of Philosophical Research, forthcoming.

5. Parent, "Privacy, Morality, and the Law," p. 269. 6. W. A. Parent, "Recent Work on the Concept of Privacy," American Philosophical

Quarterly, 20 (October 1983), 341-56, p. 346. Of course, the immorality of such privacy violation will not necessarily translate into legal sanction against it; privacy-loss on the part of especially sensitive people will often fall under that broad expanse of issues that, while

186 TRAVIS DUMSDAY

morally relevant; cannot be covered by law. It may be immoral to hold up a metre stick against a sensitive short person while his back is tumed, but such an action could hardly be outlawed. A concem has also been raised as to whether height is a fitting example of a sensitive fact, given that it is generally so readily apparent to anyone. However, I do not think this defuses Parent's example. One need only think of certain Hollywood stars who wear special risers in their shoes in order to give themselves the appearance of being a bit taller than they really are. For such people, to have a member of the paparazzi reveal this fact would constitute a loss of privacy. My thanks to an anonymous referee fbr bringing this up.

7. I am aware that the physical building should probably not be considered the actual institution here; something more is likely required. But for simplicity's sake, I use the physical mosque as a place-holder for the larger institution, of which it forms a vital part.

8. The same relation holds between individual and institutional privacy. 9. Larry May, The Morality of Groups: Collective Responsibility, Group-Based Harm,

and Corporate Rights, (Notre Dame, IN: University of Notre Dame Press, 1987), pp. 29-30. May's definition is overly restrictive, but this misstep should not affect the argu- ment made here.

10. My use of 'undergoing' rather than 'experiencing' is deliberate, as 'experiencing' might be thought to imply some awareness on the part of those losing privacy, an aware- ness which does ¡not exist here.

11. Davis, "Privacy, Rights, and Moral Value," (cited in n. 4, above), p. 120. 12. My concem here is that anonymity does not always guarantee privacy, at least with

respect to non-identifying information. One's privacy with regard to various forms of sen- sitive information can be lost even though one remains unidentified, i.e., even though one's anonymity is preserved. Sensitive information about you can be revealed without it being revealed to be about >'o« specifically, and this can still constitute a loss of privacy. Consider the following example: person Y is at a Hallowe'en party where everyone is costumed and Y has no idea who is who. In fact, Y is a visitor to the city, and does not know anyone any- way. During the course ofthe evening, Y sees Zorro and the Catwoman having an intimate conversation, and, voyeur that he is, hides behind a piece of fiimiture and eavesdrops on what they are saying. These two individuals have lost privacy with respect to their con- versation and Y, even though they remain anonymous.

13. Gavison, "Privacy and the Limits ofthe Law," p. 431. 14. See Carl Wellman,Aea/A/;gAtó (Oxford: Oxford University Press, 1995), pp. 157-70.

He takes it that agency is a necessary condition for being a rights-holder, and since groups lack agency, they lack rights. See, also, Jan Narveson, "Collective Rights?," Canadian Journal of Law and Jurisprudence, 4 (1991), 329-45.

15. My view is that the notion of a group is essentially ordinal. Unlike May, I would take any relatiori between things to constitute a legitimate denomination of 'group', but would add that, depending on the number and nature ofthe relations, a group can be more or less cohesive,;and thus the appellation takes on a stronger or weaker sense. The assem- bly of random people in a public square constitutes a group, but an assembly of devoted communists on à public march constitutes a group in a stronger sense. It is to be doubted whether a weak group can possess and lose privacy, but this question is not really relevant for our concems; A gathering of Muslims for communal worship can certainly be consid- ered to form a group in the relevant, robust sense.

16. Clearly some communal religious services would not be; the ancient mystery cults, with their secret rituals, would obviously not be, nor would their modem imitations.