The case of the camera in the kitchen: Surveillance, privacy, sanctions, and governance


  • Edna Ullmann-Margalit

    Corresponding author
    1. The Hebrew University of Jerusalem, Jerusalem, Israel
      Edna Ullmann-Margalit, The Hebrew University of Jerusalem, Givat Ram, Jerusalem 91904, Israel. Email:
    Search for more papers by this author

Edna Ullmann-Margalit, The Hebrew University of Jerusalem, Givat Ram, Jerusalem 91904, Israel. Email:


In the summer of 2007, a member of the Rationality Center at the Hebrew University of Jerusalem took it upon himself to install a closed-circuit TV camera in the Center’s kitchen. An email explained that the camera was installed in an effort to solve the problem of cleanness in the kitchen. The camera was removed a week later: within this week, the members of the Center exchanged close to 120 emails among themselves, expressing their opinions for and against the camera, and discussing related issues. Taking off from this exchange, this article explores some of the surprisingly rich set of normative concerns touched upon by the kitchen camera incident. Among them: public surveillance and people’s polarised attitudes to it, the invasive gaze and the argument that “if you have nothing to hide you have nothing to worry about,” the efficacy of disciplining behavior through sanctions along with the problem of shaming sanctions, the notion of privacy and its arguable relevance to the kitchen case, and more. In an epilogue, I offer some reflections in the wake of the incident, connecting it to the incipient literature on regulation through observation. I find that it is precisely the smallness, concreteness, and seeming triviality of this incident that helps bring a large set of interconnected, vexing normative concerns into sharp relief.


The incident

In the summer of 2007, a senior member of the Center for the Study of Rationality at the Hebrew University of Jerusalem arranged for the installment of a closed-circuit TV camera in the Center’s kitchen. An email went to all the members, explaining that the camera was installed to solve the problem of cleanness in the kitchen.

Within minutes, a response was circulated, expressing dismay: “I find this very offensive! [--] What kind of idea is this??” This message was the first in a torrent of emails; within a week, some 120 emails were circulated, using the internal mailing list of the Center.

A week later, I, as the director of the Center, ordered the removal of the camera. I should mention that my position against the camera – as well as that of the first email writer, quoted above – was, and remained, a minority position in the Center. In what follows, I make my case against the camera. Still, I attempt to canvass the set of issues involved in an evenhanded manner.

The issues that exercised the participants in the email exchange quickly diverged and multiplied. From a mere expression of opinion for or against the camera in the kitchen, people started worrying about a wider range of concerns. Among them, the effectiveness of electronic surveillance and its morality, the difference between being watched by a person and being watched by a camera, and the methods and costs involved in confronting colleagues and in sanctioning them. Also discussed was the question of humiliation involved in cleaning after oneself as compared with cleaning after someone else, as well as the nature and status of a university research center as a community. People wondered whether, from the point of view of privacy concerns, the kitchen of such a center is closer to being a clear case of a public space (like, say, an airport) or of private space (such as one’s home).

A different set of issues related to governance. What is the appropriate process that a university research center should adopt for settling such a matter; should it be a democratic vote, letting the majority opinion prevail? If so, do all Center members get to vote or should the vote be open only to those who regularly use the kitchen? Some members felt strongly that a principled moral objection to the camera, even if voiced by a small minority of Center members, ought to be accorded effective veto power overriding the majority opinion. Others felt strongly that the intense feelings of regular kitchen users whose cleanness sensibilities are offended by the occasional dirtiness of the kitchen constitute good enough reason to go along with the idea of the camera.

The incident of the kitchen camera affords us a valuable, and in some respects hilarious, opportunity to reflect upon this set of issues. It is the smallness, concreteness, and seeming triviality of the incident that helps bring this rich set of interconnected, vexing normative concerns into sharper relief. I regard the incident as a sort of peephole through which a large painted canvas reveals itself. In this essay, I try to delineate the main contours of the intricate painting that emerges, while constantly remaining aware of and referring back to the peephole itself – namely, the kitchen camera case – that gives us the particular perspective from which this exploration is undertaken.

The extensive email exchange among the members of the Center that took place during the week between the installation of the camera and its removal, serves me as an invaluable resource. I scatter quotes from it throughout the paper, under fictitious names.

Background: The Center

The Center for the Study of Rationality is an academic research center within the Hebrew University (HU) of Jerusalem. It has about 40 members (note: “members,” in what follows, are professors, as distinct from “staff”), who come from more than 10 different departments of the university. The members are bound together by their interest in interactive, strategic behavior, and by loose adherence to the conceptual framework and methods of Game Theory.1

The Center occupies the top floor of a lovely building on campus. Its premises include office space, and public spaces such as a lecture hall, a seminar room, a common room, etc. One of the Center facilities, and a hub of activity in its own right, is a well-equipped little kitchen. Much beloved by all, it is primarily frequented for its excellent espresso machine and a more or less constant supply of cookies. The maintenance staff of the HU cleans the kitchen every morning, five days a week. Otherwise, the upkeep of a clean kitchen throughout the day is up to its users. Most users clean up after themselves, most of the time. Occasionally, however, a crisis episode occurs in the kitchen; it becomes messy and dirty, sometimes quite seriously so.

Surveillance and privacy

Cameras everywhere

The real discussion should be about why a bunch of intelligent, well-educated, and probably well-meaning people in the center of rationality (no less!) cannot run their affairs without surveillance. I find it remarkable. (Jonathan, 5 Jul 2007, 20:46)

In recent years, people have grown increasingly accustomed to public surveillance. At airports and malls, in banks and supermarkets, in trains and subways, at the entrance to office buildings and to apartment buildings, closed-circuit television cameras are everywhere. Many of our daily steps are being watched and recorded; our private lives may be reconstructed to an unprecedented degree of accuracy and detail.

What is the appropriate reaction to the circumstance of near-ubiquitous surveillance?

Here are two contrasting possibilities. First, it seems reasonable to say that if we are becoming increasingly used to being increasingly watched by cameras, why should we care if we are being watched by one more camera? Seen in this light, a monitoring camera in a university center kitchenette seems hardly a reason to get incensed. All the more so since the camera is meant to serve the indisputable good purpose of keeping the kitchen clean, and to help eradicate asocial free riding behavior.

The second, converse, reaction asserts that, given the unfortunate circumstance of almost ubiquitous cameras, we should do everything in our individual power to draw some lines wherever possible; we should fight a rearguard battle to protect as much as we can of the shrinking private space around each of us.

Thus, while used by now to being monitored in banks, department stores, or airports, many of us find the idea of surveillance cameras in the cubicles of the toilets of these very institutions a travesty, let alone in the privacy of our own bathrooms at home. (Perhaps we recoil less, however, at the idea of cameras in the toilets of the high security wards of certain jails.) How do we feel about closed-circuit televisions in the workplace? We may not have entirely clear intuitions where to locate the workplace on the spectrum that lies between the two poles, of the public space on the one hand, and the private home on the other. Our intuitions may depend in part on the size and the type of the workplace (for example, on whether it is a public or a private outfit, or on whether or not it services members of the public).

How then do we feel about the university? A university has a variety of different kinds of “spaces.” Think, for example, of classrooms, auditoria, administration buildings, cafeterias, dormitories, quads, lawns, etc. Certainly different rules apply to these different spaces. So how do we feel about a research center within the university, in particular? The relevant community here is not like a group of employees in a corporation. Neither is it like a family. It is, rather, an informal and collegiate collection of people similar, in some respects, to a social club: more a Gesellschaft than a Gemeinschaft (Tönnies 2001/1887) but not identical with either. The question of the appropriateness of surveillance – which is not security or crime-prevention oriented – within such a community is therefore rather troubling.

The type of space is perhaps not a good place to start; the purpose of the surveillance may be a better starting point. After all, the reason we accept the ever stepped up surveillance around us is that it is supposed to be in the service of enhancing our security. The protection of life and property is a purpose with which we do not wish to argue. Surveillance in a university setting, however, might serve quite different aims. Academic cameras installed by management may divulge faculty members who do not show up for office hours or are late for class, those who are hostile toward students or behave improperly in committee meetings, and so on. No self-respecting academic institution, as far as I am aware, will accept these “big brother” purposes as justifying surveillance.

A camera in an academic center’s kitchen, however, is a different story altogether. The very first email that protested the installment of a camera in the Center kitchen as “very offensive,” went on to explain: “Especially since this is meant to promote neatness, rather than to prevent theft” (Miri, 28 Jun 2007, 12:59).

The distinction between criminal and non-criminal activities seems to carry weight here. Objecting to surveillance that is meant to help catch people engaged in criminal activities, such as planting a bomb or robbing the cash register, is different from objecting to surveillance that is meant to help catch people engaged in non-criminal activities. This remains true regardless of how obnoxious these activities may be, at least in the eyes of some – whether because of their free riding nature or because of their intrinsic messiness.

Visual surveillance

Security oriented surveillance is not limited to visual monitoring. It may extend, for example, to eavesdropping and wiretapping, to mail and email scanning as well as to Internet inspection, and more. The range of the surveillance devices used in the US following the attacks on 11 September 2001 was exposed to the public at the time of the enactment of the Patriot Act, occasioning heated debates. These debates highlighted the novel types of threats to public security as well as to individual safety posed by international terrorism. At the same time, they helped articulate the sensitivities concerning the price – in terms of human and civil rights – that may have to be paid if the new challenges are to be met.2

Is visual monitoring worse or less bad than, say, telephone wiretapping, inasmuch as their perceived infringements on privacy and other rights are concerned? This is an interesting, and, I believe, understudied question but it need not detain us here. I shall, however, want to bring up questions that concern the comparison between several variations on the theme of visual surveillance.

“Little objection should be raised,” said one email writer, “if someone were hired to sit all day long in the kitchen – for two shekels a day – in order to modify the colleagues’ behavior.” (Joel, 29 Jun 2007, 10:05)

How does being watched by a person, while you prepare your coffee, relevantly compare with on-screen monitoring? In fact, visual surveillance raises many questions. What is the relevant difference between the case in which the camera only transmits its pictures to computer screens in real time, and the case in which it records the images? Related to this is the question of whom to charge with the task of monitoring the screen, or with viewing the recorded tapes. How comfortable are the Center members with assigning this task to the director (or to some other members on the director’s behalf) as compared, say, with charging the administrative staff with this responsibility? And how comfortable will he (or she or they) be with this task?

One interesting suggestion in this context that came up in the exchange was to connect the computers of all of the members of the Center to the closed-circuit television camera in the kitchen. The thinking behind this suggestion is that in this way the surveillance will be carried out by all (or by some or none) of the members themselves, but not by any “central” authority, nor by the administrative staff.

If filmed, you may have further questions. You do not quite know what the camera caught, or how it will be interpreted or by whom. You also do not know whether there will be “secondary use” of these images without your consent,3 or when, by whom, or for what purpose. Finally, what is the line separating surveillance from voyeurism, anyway?

“Reasonable expectations of privacy”

Altogether, I don’t understand how the matter of privacy applies to a common kitchen. What would one want to do there “privately?” (Isaac, 4 Jul 2007, 00:19)

An academic kitchenette is not a private space in the sense in which one’s own kitchen in one’s own home is. US law has a doctrine about “reasonable expectations of privacy,” according to which unless there exists a “reasonable” expectation that what one does or says in a certain place will not be seen or heard by someone else, a surveillance operation by law enforcement authorities in that venue does not require a warrant or some other court order. Importantly, neither the simple desire for privacy in a particular place, nor the fact that one took some steps to obtain it, entitles one reasonably to expect it. The question turns on how one establishes whether, in a given instance, one’s expectation of privacy is “reasonable.”

One criterion laid down for establishing reasonable expectations of privacy is “the degree of privacy afforded by certain buildings or places.” (The other three criteria have to do with general legal principles, with the question of whether there exists a vantage point from which anyone, not just a police officer, can see or hear what was going on, and with the sophistication, and invasiveness of the surveillance technology employed.) Thus, while reasonable at a public phone booth, the expectation of privacy is not reasonable on a public highway. Nor, it would seem, would the expectation of privacy be reasonable in an academic kitchenette.4

However, a lawful search and seizure operation by local police conducted even without a warrant for purposes of law enforcement is a far cry from surveillance activity conducted in the kitchenette of an academic center for the disciplinary purpose of catching the messy users. The latter purpose is unrelated to unlawful behavior, let alone to terrorist activities. Even if, according to traditional legal concepts, expectations of privacy are not reasonable in the Rationality Center’s kitchen, and hence strictly speaking no violation of privacy can occur when a closed-circuit television monitors the kitchen, one may still reasonably object to the surveillance on broader privacy related grounds.

One such objection will be based on the chilling and inhibiting effects of the camera; “chilling effects” is a legal term used mostly in the US to describe a situation where conduct in general, and speech in particular, is self-suppressed for fear of being penalised. (A standard example is the threat of a costly and lengthy lawsuit, which might prompt self-censorship, and have a chilling effect on free speech.)

Surveillance of any sort can create chilling effects on people’s conduct, inhibiting them from acting spontaneously and unselfconsciously. Surveillance cameras are interventionist measures: their introduction affects people’s behavior, and not always in foreseen and intended ways. Such effects are acknowledged as harmful in the political context, where free speech and other rights essential for democracy might be chilled. The harmful effects will surely be at least as obvious and immediate in an academic context, where spontaneity, jest, mutual trust, and the free flow of ideas are all-important – and in the normally friendly, informal environment of an academic kitchen, all the more so.

The grounds for another objection to the camera in the kitchen will be a recent theory, labeled “contextual integrity,” according to which it is possible for violation of privacy to occur in a CCTV-monitored kitchen, depending on certain specific contextual features of the case. This approach tries to explain why the commonly accepted theoretical approaches to privacy, developed over the years to meet traditional privacy challenges, do not yield satisfactory conclusions where public surveillance is involved; public surveillance seems to fall entirely outside the range of application of the traditional approach to privacy protection that has dominated public discussion in recent decades (see e.g. Gavison 1980).

Introduced by Nissenbaum (2004), and further developed by her and others (see, for example, The Economist 2007), the theory is designed to articulate a framework for addressing the problem of public surveillance, aiming at a theoretical account of a right to privacy as it applies to information about people. It offers a model of informational privacy in terms of contextual integrity, defined as compatibility with two types of overarching norms – regarding information appropriateness, and informational flow, or distribution. The theory posits that whether or not a particular action violates the norms is a function of several specific contextual variables. When the machinery of this theory is applied in detail to the Rationality Center kitchen, it appears to yield the conclusion that the installment of a closed-circuit television camera does constitute a violation of relevant norms; whether or not this constitutes a violation of privacy depends, in the last analysis, on one’s chosen definition of privacy.

“Nothing to hide”

I see no reason for people to object to the camera in the kitchen, UNLESS THEY HAVE SOMETHING TO HIDE. And if they have something to hide, it should not be done in the common area. (Alex, 4 Jul 2007, 12:42)

The idea that the people who object to the camera have something to hide is so preposterous that it could only cross the mind of one who thinks everything in life is a simple strategic game. (Miri, 4 Jul 2007, 18:06)

As it turns out, the argument that a person who has nothing to hide should have no problem with surveillance comes up frequently in discussions of privacy issues. It is one of the primary arguments made when balancing privacy against security. “In its most compelling form, it is an argument that the privacy interest is generally minimal to trivial, thus making the balance against security concerns a foreordained victory for security” (Solove 2007, p. 2).

In truth not only privacy in the kitchen, but also cleanness in the kitchen, are rather trivial concerns, at least in comparison with the “Global War on Terror.” Still, in non-security contexts, people come up with a number of instinctive retorts in response to the seemingly compelling “If you’ve got nothing to hide, you’ve got nothing to worry about” argument. Among the typical examples: “This is not about something I want to hide: this is about it being none of your business,” or “If you had nothing to hide you would not have curtains.”6

Of relevance to these concerns is the following question, published in the weekly column “Readers Ask” of the Israeli daily Maariv:

“I am a student in the school’s photography department. For an assigned project, one of my fellow students proposed to photograph people in an elevator, without their knowledge. I believe this would constitute a gross violation of their privacy, even though no intimate scenes are anticipated. My friend claims that as the photos are intended for an academic project only, and will not be displayed anywhere else in the world, no harm is involved. Who is right?” (Baruch 2007)

The columnist’s answer was given the large-print title BEWARE, A PHOTOGRAPH HAS A LIFE OF ITS OWN. It said, in part,

“From the moment a photo was taken it has a life of its own. It cannot be confined to an academic project. It cannot be assigned just one interpretation… Do not be mislead into thinking that the seemingly innocent photo of people in an elevator cannot start wandering into all sorts of areas and uses, some of which might involve the infringement of the privacy of the photographed persons, of their rights, of their good name, or more. Hence, refrain from carrying out this project.” (Baruch 2007)

I note that a key element in this contemplated camera-in-the-elevator project involves photographing people without their knowledge. This element does not apply in the camera-in-the-kitchen case that concerns us here. Other aspects of these two cases, however, as highlighted by the columnist’s reply, are comparable. Most likely, no reasonable expectations of privacy (in the sense explained above) pertain to either the case of the elevator or to the case of the kitchen.

Concealment pertains not only to criminal acts. People are much preoccupied with concealing wrinkles on their faces or their loss of hair. Elton John had a perfectly just complaint against the paparazzi who showed him to the world without his wig. People’s attitude to their private parts, intimate relationships, serious illnesses, or the details of their bank accounts makes them want to shield all these from anyone’s uninvited gaze.

“Something to hide” is obviously taken by most people as a much broader category than wrongful doing or something of which to be ashamed;7 and what people want to protect as “personal” goes beyond what is strictly defined as “private.” The concept of intimacy indeed presupposes concealment; where everything is known and revealed to all, intimacy is precluded.

“We don’t want to expose ourselves completely to strangers even if we don’t fear their disapproval, hostility or disgust,” says Thomas Nagel, in his insightful chapter on the larger topic of what he takes to be the importance of concealment as a condition of civilization. “The boundary between what we reveal and what we do not, and some control over that boundary, is among the most important attributes of our humanity.” (Nagel 2002, p. 4)

“The ‘nothing to hide’ argument…forces the debate to focus on its narrow understanding of privacy” (Solove 2007, p. 23). There is more to privacy than wanting to hide.

Being watched

Cameras are feared more than people because people are there and you adjust your behavior. (Rachel, 30 Jun 2007, 08:12)

What difference does it make whether we are being watched by a person or by a camera?

One difference concerns symmetry. When a person watches you, you can watch him or her back. When you know that you are being watched but cannot watch back, an asymmetrical relationship is initiated, leading to an asymmetrical power structure with the potential to engender feelings of humiliation.

The one-sided mirror – say between an employer’s room and the employees’– provides an example. Or think of that quintessential scene from war movies, where the sergeant major inspects the new recruit, moving his eyes over the rooky from top to bottom, and back and forth, while the new recruit has to fix his gaze straight ahead (Margalit 2001). The asymmetry of gaze strongly attests to the asymmetry of power. Movies by directors like Kieslowski or Egoyan come to mind too for exploring with detail and fascination the theme of voyeurism, or the uninvited asymmetrical gaze, and its ramifications.

When you are in the public space, say when you walk about in the mall or the municipal park, you know that every step you make may be watched by someone. There is no problem here. There is a problem, however, when one particular person is or may be watching your every step. This can be sinister; you are being followed, stalked, or worse.8

The problem with surveillance, then, is not the fact that everything you do may be watched by someone – but rather the subtly different fact that there may be someone who is watching everything you do. This, to me, is the heart of the matter, one of the main insights I have come away with from researching this issue: that so much hangs on this switch in the order of the quantifiers.9 The corollary problem with video monitoring is that it is basically a setup that in principle makes it possible that there should be someone who is, or may be, watching you all the time.

In addition, it is often the case that the video monitoring is equipped with a recording device as well, without warning anyone about it. Also, once captured on tape, you may be watched over and over again. All of these elements are serious aggravating circumstances that go beyond the mere asymmetry of the gaze.

The panopticon and beyond

Much attention was given in recent years to Bentham’s quaint notion of the panopticon – a type of prison building whose design allows a God-like overseer to observe all prisoners without being seen by them. Foucault, famously, used the idea of the Panopticon as a formative metaphor of the project of modernism, highlighting the total loss of individual privacy – equated with the total loss of protection from centralized systems of unwanted gaze. Indeed Foucault took the idea of the panopticon an important step further by talking about the drive to self-monitoring through the belief that one is under constant scrutiny (Wood 2003).

The subtitle of Bentham’s treatise Panopticon or the Inspection House comprises a list of establishments other than prisons, “in which Persons of any Description are to be kept under Inspection” (Bentham 1995/1787). Among them hospitals, madhouses, schools, and various work places. As the proverbial locus of privacy, the home of course provides the starkest contrast to the panopticon, and to the other “inspection houses”: home is meant to provide protection from any, and all, unwanted gaze. Where on this spectrum should we locate an academic research center within a university? Do we want to see it as a “workhouse” in Bentham’s sense, thus closer to the penitentiary end of the spectrum, or rather as closer to the home end?

I take this occasion to note that the interface between my discussion of general issues raised by the kitchen camera case on the one hand, and their relevance to the specific case of the kitchen camera on the other, is delicate, indeed sometimes tenuous. Of course, the kitchen in the Rationality Center would not have become a panopticon had the camera remained, so it may strike one as an overexaggeration to bring up the notion of the panopticon in this context. A similar feeling of disconnect may occur with regard to the discussion (above) of the chilling effects of the camera, or to the problem of shaming sanctions discussed below, and possibly to further issues as well. I therefore reiterate the peephole simile mentioned at the outset, and point out that, rather than arguing for direct relevance, I take the incident of the kitchen camera as an opportunity to reflect upon a wide set of normative issues, and suggest some normative conclusions.

Here is a different tale about a camera in another academic kitchen. It might throw a different – and helpful – light on our case.

“If I remember correctly, the first web-cam was used exactly to monitor a shared coffee room in a university. It involved computer science students who didn’t want to walk to a different floor to get coffee if the pot was empty. So they set up a camera via the Web, to be able to see whether the coffee pot was full. This didn’t cause any discussion of privacy.” (Ofer, 4 Jul 2007, 18:48)

The email message concludes: “It seems to me that the problem with the camera is not really the issue of privacy but rather of being ‘checked up on’.”

This is a correct insight. It seems as if the fact that one’s image and activities are projected onto a screen is not, in and of itself, what one objects to in general; it also seems as if the violation of one’s privacy is not what one objects to in particular. What people find objectionable in the case of the Rationality Center kitchen camera is “being checked up on.” People object, first, to the disciplinary purpose behind the installment of the camera. Secondly, people seem to be sensitive to the grotesque disproportionality of the surveillance device, relative to the purpose behind its installment.

Regulation and governance

Disciplining behavior

Confronting people who were spilling coffee and sugar or not cleaning their cups based on video records is not going to be a lovely scene. (David, 29 Jun 2007, 09:25)

Suppose the closed-circuit television is connected to the computer screens of all members of the Center, so that anyone can watch what goes on in the kitchen in real time. No “central agency” is charged with the disciplinary action; rather, the charge is equally distributed among all Center members.

This idea of peer discipline may appear attractive but its effectiveness is dubious. When it is everyone’s responsibility to catch transgressors, it ends up being the responsibility of no one. Depending on who happens to be watching their screens at the relevant time, catching the culprits becomes a matter of happenstance.

If people do not make it their habit to connect to the closed-circuit television, the solution offered by it is ineffective. It they do, and they happen to see a colleague messing up the sink, or leaving the milk outside the fridge and the foam pipe unwiped, what exactly are they supposed to do? Whatever they do, they might find themselves dealing with phenomena such as recrimination, badmouthing, rumor spreading, whistle blowing, or shaming.

All of these constitute part of the price that the Center community might have to pay for the methods it chooses to use for catching the culprits, and for punishing them. The adverse effects on the working atmosphere within the Center, and the possible damage to the values essential to the fabric of the Center community, need to be balanced vis-à-vis the value to the Center community of a clean kitchen.

Suppose, next, that a taped record is kept of all images transmitted from the camera to the monitors. Surely, the potential effectiveness of the camera for disciplinary purposes in this case increases significantly. Not only is it possible to discover who was remiss, but it is also possible to confront the wrongdoers with “proof” of their misdeeds.

However, the very existence of the taped record cannot be taken lightly. Some rules and regulations have to be decided upon ahead of time, regarding questions such as who can see the tapes, how to prevent their misuse or abuse in general – and access to them by non-approved parties in particular, how long they are to be kept, and more. Moreover, some second-order questions need to be addressed: what are the correct procedures to settle the procedural questions just posed (e.g., who decides who can see the tapes, etc.). I am not suggesting that these issues are insurmountable; but I am saying that they are delicate matters, to be addressed with tact, and with due transparency.

Let us now consider the act of confronting the transgressors with on-camera proof of their misdeeds, reminding ourselves that the transgression in question is not a criminal offense or felony. Nor is it even a disciplinary offense in the usual sense (like cheating in an exam, or plagiarism) but rather misdeeds of a different, “softer” nature.

The idea of using surveillance cameras (or polygraphs, for that matter) within the family is, to most people, unthinkable. When you find a dirty sink, or unwiped coffee stains on the floor at home, you may tell your children off, even punish them. But already when it comes to your spouse, the question of confronting him or her is not an altogether simple matter. Family politics and issues involving baggage from the past are often complicating factors (Ullmann-Margalit 2006).

When the culprits are one’s colleagues, the issues are comparably complex, perhaps even more so. In analogy to the case of the family, it may be better, and wiser, to hold back. Moreover, it may be wrong to see the act of confronting a transgressor with proof of his or her misdeeds as a triumphal act, parallel to the clinching climax of a criminal court case: we must see it, rather, in the different light of an act of shaming.

Shaming and shame sanctions

We also need a mechanism to deal with those who are “caught”. I suggest that the first step will be a discreet conversation with [the Director] or somebody. (Rachel, 30 Jun 2007, 14:25)

To treat a person who forgot to return the milk to the refrigerator as someone who is “caught” and who will, “as a first step,” be summoned to a “discreet” conversation the Director, is mind boggling. (David, 30 Jun 2007, 21:59)

Shame is a private emotion. I am ashamed when I realize that what I said or did diminishes me in my own eyes; when I privately experience embarrassment and loss of self-respect (Williams 1993, p. 89). Shaming is a social act; one person putting another person to shame. It can also be public, when the act of shaming occurs in the presence of an audience, or of witnesses. Shame induced by a second party intrinsically involves elements of derision and contempt, and hence loss of dignity and humiliation (Nussbaum 2004; Margalit 1996).

In the Jewish sources, “whitening” someone’s face in public, namely shaming them, is likened to spilling that person’s blood. The commandment not to shame is comparable in its importance to “Thou shall not kill.”

The theme of shame sanctions has gained growing presence in the legal literature since the mid-1990s (Kahan 1996; Massaro 1997; Posner 2000). Shame sanctions have a strong element of spectacle for the spectators, and shame for the offender. Their effects on the public can be politically dangerous; their effects on the offender violate human dignity.

Public punishments (whipping, flogging, dunking, branding) used to be an integral part of the old punishment traditions but have largely faded away. The current US practice “takes milder forms, such as requiring offenders to wear shirts describing their crimes, publishing the names of prostitutes’ johns, or making offenders sit outside public courthouses wearing placards” (e.g. “I am a Drunk Driver”) (Whitman 1998, p. 1056).

Caught by a closed-circuit television camera, offenders are exposed red-handed, or actually dirty-handed. The exposure here is quite literal: one is seen misbehaving; perhaps one’s misconduct is even broadcast on tape in the presence of others. The very exposure is meant to be, and to do, the condemnation.

Informal shaming sanctions may comprise scolding and rebuking, and also ridicule, contempt, avoidance, shunning, and more. Of their several idiosyncratic features, one is that people’s reaction to them is extreme: either one finds them easy to ignore, or one finds them particularly harsh, and “cripplingly diminishing of self-esteem” (Braudway 2004, p. 80). The effectiveness of such sanctions on people of the first sort is nil; with respect to the second sort, the effect is likely to be overkill, and hard to predict (Whitman 1998).10

Also, shaming sanctions seem particularly vulnerable to the phenomenon of diminishing effects. So, while the administering of a shaming sanction for the first time may have overkill effect, with consequences that are difficult to control and predict, its repeated use may prove ineffectual (Harel & Klement 2007).

Stigmatization imposes costs on the enforcers too. In order for it to work, stigma relies on the active cooperation of individuals who must incur costs in privately sanctioning the offenders; the effectiveness of private sanctions is based on the willingness of individuals to incur such costs. “Other things being equal, the larger the costs private enforcers incur in the imposition of private sanctions, the less the willingness of private enforcers to stigmatize, and consequently, the less effective stigma becomes.” (Harel & Klement 2007, p. 358) Assessing the costs borne by private enforcers is crucial therefore for predicting the effectiveness of the proposed scheme of stigmatization.

An academic center like the Rationality Center, as a community, must face the consequences for its morale and esprit de corps once it becomes a surveillance community. It must face the possibility that some of its members will become peeved, even humiliated, by the effects of shaming.11 It must also be concerned with the question of whom to charge with delivering the rebuke, whether this charge is fair, and what its costs might be. (And, how about the idea of a camera recording the sanctioning proceedings too, in the spirit of “guarding the guardians?”)

Deterrence and its limits

Game theory teaches us that one can modify people’s behavior with the right incentives. (Alex, 6 Jul 2007, 10:51)

We put the camera, and if that does not work, then I suggest a high voltage fence as the next step… (Andre, 3 Jul 2007, 10:23)

The camera as such does not clean the kitchen. But can it guarantee a clean kitchen? Evidently some in the Center not only hoped that it will, but were convinced that it will. Why?

One line of thinking is, surely, that the camera deters potential transgressors from transgressing. The complementary line of thinking is, presumably, that if transgression does occur, then the camera enables catching of the culprits. Once the culprits are caught and punished – and the news goes around – deterrence kicks in again and transgression stops.

Deterrence can work – up to a point. It is clear that sustained cleanness in the kitchen cannot rely on the camera’s deterrence effect alone. Incidentally, for partial deterrence the camera may not in fact be necessary. As one contributor to the email exchange pointed out, “in some experiments by behavioral economists, merely having a poster of a face with big eyes on the wall markedly improved compliance.”12

So much for deterrence. Let us now probe the complementary argument, about catching the culprits. What is the causal chain that supposedly connects the catching of the culprits with a forever clean kitchen? With what degree of confidence can we expect people’s behavior to improve as result of disciplining measures?

Changing people’s behavior is a notoriously complex business. Much of social science – and most of education – is about this. Psychologists and priests, criminologists and political activist, sales persons, and advertising agencies – all of these and many others attempt to affect and change people’s behavior. The methods they come up with range widely over persuasion, propaganda, brainwashing, coercion, conditioning, manipulation, formal and informal sanctions, incentives and disincentives, carrots and sticks. In an open society, the choice may be somewhat restricted.

Confronting a person with evidence of his wrongdoing, and punishing him for messing up the kitchen may have the desired effect of “teaching him a lesson.” But then, it may not. Or it may have the desired effect for a while and then wear off. When his colleagues hear about this, some may indeed improve their behavior in the kitchen. But then, some may not.

Why do some people react one way, and others the other? Under what conditions is one reaction likelier than the other one, and by how much? This is the stuff of complex social science. The puzzling point to me was realizing that for several of the participants in the exchange, the full success of the surveillance device was a sure thing. In their minds, the installment of a camera amounted – as a matter of necessary truth and logical certainty – to the achievement of 100% cleanness in the kitchen. Consider: “The dichotomy is not a false dichotomy; it’s a real, practical one. If we want to keep the kitchen open —AND I DO—we either have the camera, or the administrative staff keeps cleaning it” (Isaac, 5 Jul 2007, 17:46.)13

In fact, the unbearable lightness of accepting this means–end connection was characteristic of the proponents of the camera, who saw the chain as an ultimate winning argument. Strikingly, this was not the case for the camera opponents, who had qualms with the normative aspects of the means–end chain. Here is a particularly telling comment, revealing how people’s views regarding the anticipated efficacy of the camera device intertwine with people’s views about its normative acceptability: “I suspect that the people who think this measure [i.e., the camera] is repugnant also don’t think it will work, and the people who think it WILL work, don’t find it repugnant” (Miri, 30 Jun 2007, 17:17).

One factor the proponents of the camera did not consider, but we surely must, is that the presence of the camera may start altogether unanticipated chain reactions, bringing about unintended consequences. For example, people may wish to avoid any encounters with the camera, and hence leave their dirty coffee cups all over the place rather than return them to the kitchen – thus raising the general messiness level at the Center, not reducing it.

The point is that people do not always respond in the way that somebody intended them to respond. Their reaction to the camera may diverge from the causal chain meant to improve their cleanness behavior in the kitchen. The design of human response to a novel technical device has its limits; sometimes it even backfires.

An attempt by a respected Los Angeles medical doctor to improve his colleagues’ cleanness related behavior was reported not long ago by Stephen J. Dubner and Steven D. Levitt (of Freakonomics fame). Many medical studies have shown, they say, that hospital doctors wash or disinfect their hands “in fewer than half the instances they should” (Dubner & Levitt 2006).

The story highlights how hard it is to change people’s entrenched habits, and how much effort can be required to solve the seemingly simple problem of changing people’s hand washing behavior.

In contrast to the hospital case, cleanness in the Center kitchen serves no distinct instrumental purpose. Nobody is going to die because of a substandard level of cleanness in the kitchen. (Besides, the following morning the cleaner arrives.) In the hospital, a policy of zero tolerance toward transgressors is instrumentally justified; not so in the case of an academic kitchen. Bluntly put, I believe that zero tolerance toward cleanness offenders is intolerable within the context of an academic community. True, habitual free riders and offenders should not be tolerated, but occasional lapses and instances of absentmindedness should. A machinery of spying, catching, confronting, shaming, and sanctioning – all in the name of reaching 100% cleanness in the kitchen – is in my view unacceptable.


Let me suggest to conclude this lengthy and sometime heated debate by casting a vote (electronically) among the Center’s members. (Benny, 5 Jul 2007, 12:22)

Who should have made the decision about the kitchen camera, and how?

The way it actually happened was that a senior Center member, who had entered the kitchen and found it in a particularly disgusting mess, felt that he had finally “had it.”14 Set on an energetic problem solving course, he went ahead and installed the camera, technically aided by two administrative staff persons. The office then immediately sent an email saying, “For your information we have installed a closed-circuit camera in the Center’s kitchen. Please do your part to keep the kitchen tidy. Thank you for your cooperation.”

In real time, Center members were not aware of what had actually taken place. Indeed, they probably imagined it quite differently. An early contribution to the email exchange expressed the following opinion: “We have a highly effective administration/staff team…If they have decided to put a camera in the kitchen, I am sure it was done after serious internal deliberation, and therefore we should all (and I do) support it” (Eli, 1 Jul 2007, 14:16). Several members expressed instant support for this position. Another put it thus: “I trust the management. If they felt that the situation calls for such an arrangement, which is not trivial, I trust them that the problem is real. The fact of the matter is that most places do not reach such a solution.” (Rachel, 5 Jul 2007, 15:13)

It appears, then, that at least for some members it was perfectly acceptable that the management of the Center (director, executive committee, or senior staff) should be making such a decision. Other members wondered whether the Center has any authority to install a camera in the first place, and whether the university administration should not have been consulted about the matter.15 A democratic alternative to both the executive and the administration options is that the Center-wide community should decide the matter. Namely, first, the members should have reached a general agreement among themselves that the cleanness of the Center kitchen is indeed a problem they, as a collective, want solved; second, they should have agreed upon a mechanism for arriving at a solution; and only thereafter should they have gone ahead and implemented the agreed upon solution. “Agreed upon” might involve any number of techniques: majority vote, forming a kitchen committee to present options that would be voted on later, designating a special “kitchen czar” invested with full authority to solve the problem as he or she sees fit, assigning the responsibility to the Center director, etc.

Presented with a similar problem, different communities, and different workplaces would find either the executive option or the democratic option (in one of its versions) more suitable. The question of which option fits an academic community such as the Rationality Center is, in my view, nontrivial. Basing himself on proper governance principles and theories, at least one Center member was very clear that the answer to this question must be the executive:

A sub-unit of an organization is not a democracy, and its director need not abide by a majority rule… The only duty of a director is to run his or her unit well, sometimes against the wishes of the majority… The Center is not a pirate ship owned by its members… The university governance [has] clear and continuous chains of command… Since you [namely me, EUM] were nominated by the Hebrew University administration to direct the Center, you are, indeed, authorized to decide on the camera issue, unless overruled by someone higher up the HU administrative hierarchy. (Joel, 6 Jul 2007, 00:42; 13:00)16

Others felt that the democratic option was more suitable for the Center. But here, too, there was no agreement as to whether all should be assigned equal weights. “Those who hold offices in [the building]should be assigned more votes than those who do not; four-to-one, say? Let’s debate now this ratio, take the vote, and finalize” (Joel, 5 Jul 2007, 18:28). Another member suggested a secret ballot, “to compensate for the status quo effect: the camera has been installed (and therefore this is the status quo now)” (Yoav, 5 Jul 2007, 14:42). Reacting to these messages, one member quipped: “I enthusiastically abstain” (Emanuel, 5 Jul 2007, 15:39).

If the executive option prevails, then the director (or a body on the director’s behalf) decides one way or the other, and this settles the matter. But let us suppose democracy is opted for, and a general discussion takes place among Center members about the pros and cons of the camera in the kitchen.

Clearly, if the outcome of the discussion is that aversion to electronic self-surveillance is overwhelmingly stronger than the desire to catch those who violate the cleanness norm, then the camera solution will be rejected. Judging from the contributions to the email exchange, however, I am not clear that this would have been the probable result in the case of the Rationality Center, had such a general discussion taken place. What happens, however, in the opposite case, namely if there is a majority in favor of the camera, but no unanimity? In particular, what happens if the minority who object to the camera feel very strongly about the issue, basing their objection on deeply held convictions and principles? Should the majority simply overrule this minority? Is this case to be decided by the counting of votes, or is this a case in which veto power might be granted to the opposition, no matter how small?

I find this question troubling. It is troubling to me partly because I suspect that, had my own opinion been pro-camera, I might probably not have seen the point of even raising this question in the first place, and partly because I do not have a proper grip on the arguments that might justify vetoing rather than voting. As citizens, most of us take it for granted that the polity is to go by majority rule even about issues which, to some segments of the citizenry, pertain to defining core convictions (for example, abortion in the US, or sovereignty over Jerusalem in Israel). However, regarding smaller communities, semi-formal, and voluntary associations, this might be less clear-cut.

How do we think in this connection about a small, semi-formal, and voluntary academic community, organized as a university research center? In all likelihood, layers of common background beliefs and shared worldviews bind such a community together, transcending differences along political or religious dimensions. Also, in all likelihood such a community is only rarely called upon to take a decision that touches upon the deepest convictions of its members. So on the rare occasion that a community like this must make such a decision, is it to be taken for granted that majority rule prevails? Is the ardent and deeply felt desire, on the part of the majority of members of an academic community, to be able to make their coffee in a clean kitchen comparable to the deeply held desire on the part of the minority, to avoid turning the community into a surveillance community? I leave this question hanging in the air, because I do not know how to ground it.17


I don’t think that Rabelais or Swift or Waugh could have invented something as hilarious as the discussion that took place here. Keep the good work, folks. (Jonathan, 1 Jul 2007, 10:00)

What are my lessons from this incident?

One lesson has to do with the notion of solution. As noted by one email, “A dirty kitchen is disgusting. A camera-surveyed kitchen is repugnant. On balance, I am not sure the solution is better than the problem” (Miri, 3 Jul 2007, 16:03). Reflecting on the matter, I realize that a lot hangs on what one means by a solution here. Is it appropriate to strive for a solution that guarantees 100% cleanness, such that no dirty episode occurs in the Center kitchen, ever?

If we focus on the notion of the “problem” rather than on the “solution,” perhaps we shall come to realize that this may be one of those cases (familiar to clinical therapists), in which a solution – or a resolution – is achieved largely by learning to accommodate to, and to live with, the problem. Namely, perhaps the solution here lies with teaching ourselves to live with somewhat lower standards of cleanness than what some of us expect at home, and with wiping up after others, occasionally. In any case, we can comfort ourselves with the thought that by the next morning the place will be clean again.

A second lesson derives from a striking observation about the email exchange. It is that most Center members made up their minds about the issue instinctively and instantaneously; their instincts were made up before their minds were. This contrasts with the attitude of many people to moral dilemmas, where they find themselves agonizing long and hard about what their opinion should be.

The corollary observation is that most of the participants in the exchange felt that the natural light of obviousness is on their side; they seemed not to recognize the potential validity, or even legitimacy, of the opposite attitude. People on each side of the argument tended to see the alternative to their own view as ridiculous, disingenuous, and even perverse.

So I gained the insight that cases of public surveillance are troublesome because they seem to “drive opponents into seemingly irreconcilable stances” (Nissenbaum 2004, p. 101). The new methods of gathering information drive some people into indignation, while others remain unconvinced and even puzzled by what they consider a mere dislike of new technologies and practices. It seems that traditional theoretical frameworks fail to handle these conflicting attitudes and stances.

Having said that, I should mention incipient work proposing novel uses of a variety of new technologies – including, but not limited to, surveillance – to help constrain and regulate antisocial behavior. As an example, consider the program launched in Cincinnati, whereby spectators can report hooligan behavior of football fans in the local Bengals stadium by calling a certain hotline telephone number. The behavior of the suspect fans is then monitored with the help of the stadium’s CCTV cameras, making it possible for security officials to remove, and even arrest, the misbehaving individuals. “Given the problems associated with soccer hooliganism around the globe, the innovation deserves serious attention.”18

The idea here is to combine the mechanism of individual reporting with the use of surveillance cameras, to achieve control of undesirable behavior. As such, it fits in with the budding body of literature concerned with exploring innovative ideas for the use of information aggregation technologies for the purpose of deterring, detecting, and, where needed, punishing people’s misconduct. A good example of this is the proposal by Strahilevitz (2006) to put to use “How’s My Driving”–style programs in a variety of social contexts. Among them, in addition to the area of traffic regulation, he mentions the behavior of soldiers, police officers, hotel guests, sports spectators, and participants in virtual worlds.

In all of these, the suggestion is that the applications of reputation tracking systems will help “transform loose-knit environments, where reputation often fails to constrain antisocial behavior, into close-knit environments, where reputation constrains misbehavior more effectively” (Strahilevitz 2006, p. 1699). Intriguingly, however, I note that the kitchen camera case demonstrates that close-knit environments do not guarantee the success of constraining misbehavior: indeed, it shows that precisely because a close-knit environment is involved, in which the reputation of participants counts, the problems of shaming sanctions become worrisome.

Innovative and fascinating, this new set of ideas merely begins to scratch the surface of the host of issues that come together to form a heterogeneous field of study that concerns itself with what might be termed “regulation through observation.”19 At issue is the intricate, and as-yet intractable, interface between a number of disparate concerns – with surveillance, control, privacy, shaming, and social norms. In a way, the camera in the kitchen incident serves as a paradigmatic case study – a microcosm of sorts – of this study area. Its triviality notwithstanding, this case attests to the fact that regulation by observation touches some basic chord in the minds of people. This chord, moreover, seems connected to a variety of key sensibilities that are involved in determining people’s essential life choices and core convictions about broad social issues.


In the months following the Kitchen Camera incident, I had occasions to tell various people the story of the “Curious Incident of the Camera in the Kitchen.” The idea of an academic community acting as its own “big brother” in the name of kitchen cleanliness variously regales, and appalls, the listeners. In addition, I am invariably asked how clean the kitchen now is. The answer is that, well, overall it has been kept “reasonably clean, thank you very much.”

This is of course partly attributable to the effect of the email exchange itself, which helped heighten people’s awareness of their personal responsibilities in the kitchen. Also, a number of additional measures were taken. For example, the doorstopper was removed, so that the kitchen door now slams shut, and can only be opened with a key, thus reducing the chance that unauthorized persons will use the kitchen, and, again, serving as a reminder.

Finally, in a brilliant move, one Center member put up a big sign next to the electric fan on the kitchen wall (where the camera was), proclaiming “CLOSED-CIRCUIT fan IN OPERATION.” In doing so, he succeeded in artfully reminding people of the note about the camera that was there before, and thus, vicariously, in gently nudging them to clean up. He thereby also succeeded in producing some good-natured smiles – which may well be the appropriate response to the Curious Incident of the Camera in the Kitchen anyway.


Gil Kalai assisted me in writing this article. I wish to acknowledge my large indebtedness to him. I also want to thank Maya Bar-Hillel, Michael Biggs, Avishai Margalit, and Cass R. Sunstein for helpful comments on earlier drafts.


  • 1

    For more about the Center, see:

  • 2

    For more see Nissenbaum (2004), p. 113.

  • 3

    “Secondary use” refers to the use of data obtained for one purpose for a different and unrelated purpose without the person’s consent.

  • 4

    This doctrine was developed in connection with the Fourth Amendment’s protections (enacted in 1791) against “unreasonable searches and seizures.” For more see LaFave (1996), Nissenbaum (2004, pp. 112, 117–118), Katz v. United States, 389 U.S. 347, 360–361 (1967), and also (New York Surveillance Camera Players).

  • 5

    How about the privacy of a suicide attempt in public? In a landmark 2003 decision, the European Court of Human Rights has ruled that a British man’s right to respect for his private life was violated when closed-circuit television footage of him attempting suicide was released to the media. The court awarded him damages of £7,800. (Bingham 2003).

  • 6

    Some guide books to Amsterdam, in their section about the famous Red Light District, talk of Calvinist open-curtain culture to show the world that nothing untoward is happening behind their front doors; the architecture of many Dutch houses indeed allows for an uninterrupted view from front to back.

  • 7

    For an opposite view, consider Posner’s, according to which privacy consists of a person’s right to conceal discreditable facts about oneself (Posner 2007/1973).

  • 8

    Consider the lyrics of the well-known song of the “Police”: “Every breath you take /Every move you make /Every bond you break /Every step you take /I’ll be watching you” (Sting 1983). The beauty of the song notwithstanding, the situation it describes is uncomfortably close to stalking.

  • 9

    The benign case is “For every step s that you make, there is, or there may be, person P, such that P is watching s.” The sinister case, in contrast, is “There is, or there may be, person P such that, for every step s that you make, P is watching s.”

  • 10

    Kahan and Posner (1999, p. 386) also make the point that the stigmatizing effects may often be too large or too small, and thus they do not render themselves easily to achieving the desired results.

  • 11

    There is, however, a different approach, according to which not all shaming is bad. This approach (Braithwaite 2000; Ahmed et al. 2001) contrasts stigmatization, conceived as disrespectful shaming, with reintegrative shaming, where society’s disapproval of the offender is communicated within a framework of respect for the offender. Shaming done within a cultural context of respect for the offender can, according to this approach, be both an efficient and a just form of social control, provided the social conditions for such shaming are successfully identified and implemented.

  • 12

    Bateson, Nettle and Roberts (2006) report an elegant study that backs up this statement, with regard to norms of generosity.

  • 13

    Note, too, that the causal (not logical) means-end connection here is not free of potential obstacles and glitches of a technical nature. We might think, for example, of the possibility that the view of the camera is blocked, supposing two or more people are in the kitchen at the same time, or that someone obstructs the view on purpose. Again, the quality of the tape might be bad, or nobody is monitoring the closed-circuit television when a transgression occurs (assuming no recording takes place), and so on.

  • 14

    “I have had it with people leaving the milk out of the fridge, using half a liter of milk to make foam for one cup, and then leaving the rest to spoil and stink there, spilling coffee and sugar everywhere and not cleaning, not washing their cups, throwing garbage on the floor, … and the list just goes on and on” (Alex, 28 Jun 2007, 23:06).

  • 15

    In the wake of the kitchen camera incidence, a Center member sent a letter to the president, and to the legal adviser of the HU, asking them what the HU policy is regarding the installation of CCTV by individual units of the university, if indeed such a policy exists. Ten days later a memorandum was circulated by the Secretary to the Administration, addressed to all schools, departments, and units, announcing that it is strictly forbidden to install cameras within the premises of the university, and that any request to do so requires appropriate coordination and special permission.

  • 16

    Intriguingly, the person who wrote this had suggested, in an earlier message, to take a vote. He prefaced the quoted passages with the comment/apology: “Indeed, my previous call for a vote was a mistake, inconsistent with what I have been teaching in the last few years.” For more on the pirate ship analogy, see Casey (1992).

  • 17

    Consider a different case where the question of majority rule may be troubling: a group of people is traveling in a car. Some find the air in the car stifling and want the window open for fresh air, the others get chilled when the window is open and want it kept closed. What is to be done?

  • 18

    Reported by Brunsman (cited in Strahilevitz 2006, p. 1763).

  • 19

    I am indebted to Cass R. Sunstein for this label. It echoes, of course, Foucault’s “Surveiller et punir.”