Censorship and Access to Expression

Censorship and Access to Expression1 [Penultimate draft. Please cite published version in Handbook of Information and Computer Ethics. Kenneth E. Himma and Herman T. Tavani Editors. John Wiley and Sons: New York, 2008.] Kay Mathiesen School of Information Resources and Library Science University of Arizona Introduction No one wants to be a censor. Or, more precisely, no one wants to be called a “censor.” To describe a person as a censor, or an act as one of “censorship,” is to condemn the person or the action. While there are numerous calls to arms to resist censorship and compilations of instances of censorship across the globe, little work has been done to help us understand the concept itself. Without getting clear on what we mean by censorship, it is difficult to get a grip on exactly what is wrong with it, and indeed, on whether it always is wrong. Those who use the term simply assume that we know what it is, that it is wrong, and do not look much further. One might hope that work in law or philosophy might provide the answer. However, the concept of “censorship” is commonly used in ways that go much beyond the strict confines of First Amendment law.2 Nor have philosophers, even those who have written much of “freedom of expression,” tried to provide a conceptual analysis of 1 Ken Himma deserves many thanks for his extensive critical comments, which pushed me to clarify and defend my definition of censorship. Thanks for many helpful discussions are also owed to Don Fallis and to my students in the School of Information Resources and Library Science at University of Arizona. 2 I will argue in section three of this paper that we ought not to equate censorship with infringement of first amendment rights. In other words, I argue that censorship is not be definition an act of the government. 1 Electronic copy available at: http://ssrn.com/abstract=1264451 censorship itself.3 In this chapter, I try to fill in this gap by providing a definition of censorship. With this definition in hand, I consider the sorts of justifications given for censorship and canvass the arguments against censorship. I. The Interest in Access to Expression Censorship limits access to an expression, either by deterring the speaker from speaking or the hearer from receiving such speech.4 By an “expression” I mean anything that may be composed by one person and communicated to another. This includes such things as speeches, personal communications, books, articles, compilations of data, art works, photographs, and music. Given that censorship limits access to expression, it is important to have clearly before us why access to expressions is valuable. Stuart Cohen (1993) provides an admirably clear and convincing account of the fundamental human interests that can only be satisfied if there is access to the expressions of others. Cohen links our concern with freedom of speech to three fundamental interests: (1) the interest in expression, (2) the interest in deliberation, and (3) the interest in information (223230). Cohen defines the interest in expression as, “a direct interest in articulating thoughts, attitudes, and feelings on matters of personal or broader human concern and perhaps through that articulation influencing the thought and conduct of others” (1993, 224). Note that, while Cohen’s emphasis is on acts of expression directed to others with 3 See e.g., Brink 2001, Cohen 1993, Dworkin 1981, Scanlon 1972, Scoccia 1996, and van Mill 2002. 4 This characterization will be made more precise and defended in the following section. 2 Electronic copy available at: http://ssrn.com/abstract=1264451 the goal of “influencing thought and conduct,” this should not be understood as limited to those works that are clearly propositional in character. Works of art, such as novels, music, photographs, and paintings also “articulate” “thoughts, attitudes, and feelings.” There are a number of ways in which access to expression supports this interest in expressing. First, as Cohen notes, most acts of expression are acts of communication to others.5 By promoting access to information, we are enabling the success of such expressive acts by connecting, for instance, the writer with the reader. Second, in order to engage in acts of expression people need a rich information culture that will allow them to develop their ideas and learn how to communicate them effectively. There is, however, more to the interest in expression than Cohen’s account covers. Cohen does point out how access to expression satisfies the receiver’s deliberative and informational interests, but he does not recognize an independent interest in accessing expression. It would be a mistake, however, to think that the interest in accessing other’s expressive acts is merely derived from the more fundamental interest that others have in expressing themselves to us. Human beings have an independent interest in accessing the expressions of others. Just as we have a need to express ourselves, we have a need to hear others expressions. We have, in other words, “a direct interest in accessing the thoughts, attitudes, and feelings of others on matters of personal or broader human concern.” In addition to our direct interest in expressing ourselves and hearing what others have to say, access to expressions are necessary to satisfy what Cohen calls our 5 Not all expressions are acts of communication to another person; think, for example, of a private diary (though one might argue that it is an act of communication with one’s future self). 3 “deliberative interests.” The deliberative interest concerns our ability to revise and gain a deeper understanding of our individual and collectively held beliefs and commitments. This requires access to expressions of others, due to “the familiar fact that reflection on matters of human concern cannot be pursued in isolation. As John Stuart Mill emphasized, reflection characteristically proceeds against the background of an articulation of alternative views by other people” (Cohen 1993, 229). It is only in the context of free access to the full range of “alternative views,” according to Cohen, that we can engage in deliberation on what to believe, value, and do.6 Finally, access to other’s expressions allows us to leverage our collective epistemic labor and satisfy what Cohen calls our “informational interests.” The “informational interest” is the “fundamental interest in securing reliable information about the conditions required for pursuing one’s aims and aspirations” (229). Without access to such information, individuals and groups will be unable to effectively carry out their aims. In a free society, we assume that the individual and the collective good is promoted by persons having the ability to determine for themselves what they value and having the freedom to pursue those goals effectively (as long as they do not interfere with a similar pursuit by others). Access to the information and knowledge contained in the expressions of others allows us to do this. Furthermore, the wellbeing of both individuals and groups requires that we base our actions on the best information available, information we are unlikely to be able to get all on our own. To summarize, we have an interest in access to expressions based on our fundamental interests in communicating with others, both as speakers and as hearers. In 6 Note again that such alternative views are not only communicated through works of opinion, but also through works of art, novels, and other creative works. 4 addition, we have a fundamental interest in accessing expressions based on individual and collective deliberative and informational interests. Notice that the focus here is on the primary importance of the capacity of persons to communicate with each other, rather than on the interest in the freedom of the speaker to speak. Standard discussions of free expression focus almost exclusively on the importance of person’s freedom to engage in acts of expression. However, it is less often noted that such expressions are fundamentally acts of communication. If persons do not have the right to receive information so expressed, the act of communication is prohibited. The view advocated here is that the goal of freedom of expression is not that a speaker gets to speak, but that people are able to communicate with each other. The goal of speech is to reach a willing hearer. Indeed, often we are concerned with protecting freedom of speech for the sake of the hearer, rather than for the sake of the speaker. For example, the U.S. Constitution’s First Amendment protection of freedom of the press is not there to protect the members of the press’s ability to write stories, but to protect the right of the public to be informed by such stories. As the U.S. Supreme Court put it in Griswold v. Connecticut, “The right of freedom of speech and press includes not only the right to utter or to print, but the right to distribute, the right to receive, the right to read (Martin v. Struthers, 319 U.S. 141, 143 ) and freedom of inquiry, freedom of thought, and freedom to teach (see Wieman v. Updegraff, 344 U.S. 183, 195 ).”7 7 In a concurring opinion to Lamont v. Postmaster General, Brennan expanded on this theme, writing: “It is true that the First Amendment contains no specific guarantee of access to publications. However, the Protection of the Bill of Rights goes beyond the specific guarantees to protect from congressional abridgement those equally fundamental personal rights necessary to make the express guarantees fully meaningful…. I think the right to receive publications is such a fundamental right. The dissemination of ideas can accomplish nothing if otherwise willing addressees are not free to receive and consider 5 Freedom of speech, thus, includes the liberty to express one’s point of view to others and the liberty to receive any such expressions. To censor is to interfere with this liberty by either suppressing such expressions or preventing the reception of such expressions. In short, a censor wishes to prevent a willing speaker from speaking to a willing hearer. To censor is to interfere in an act of communication between consenting individuals. II. Defining Censorship Above I characterized censorship as limiting access to content, either by deterring the speaker from speaking or the hearer from receiving the speech. Or, more informally it is to interfere with acts of communication between consenting adults. In this section, I develop and defend the following more precise definition of censorship: To censor is to restrict or limit access to an expression, portion of an expression, or category of expression, which has been made public by its author, based on the belief that it will be a bad thing if people access the content of that expression. Before considering some possible objections, it is worth explaining some key features of this definition. First, in order to leave it an open question whether censorship can be justified, I avoid defining censorship in a way that would beg the question as to whether it is always wrong. In other words, this definition tries to capture only the descriptive, and not the normative element of censorship. Some might object that this fails to recognize the fact I noted above; we typically use the term “censorship” to pick out wrongful attempts to suppress expressions. And, to the extent that suppressing an them. It would be a barren marketplace of ideas that had only sellers and no buyers.” (U.S. at 308) 6 expression is not wrong, we are often loathe to call it censorship. So, for example, one might think that it is not censorship to try to restrict people’s access to child pornography. My view is that significant progress can be made by simply focusing on the descriptive element of what makes something a candidate for “censorship.” Recently, Thomas Carson (2006) defended a similar approach to the definition of lying. Carson argues that, “There are good pragmatic reasons for us to use the concept of lying to help point out and distinguish between salient features of actions and thereby assist us in making moral judgments. In order to serve this purpose, the concept of lying must be defined independently of controversial moral assumptions” (288). The same reasons for avoiding normative elements apply to the definition of censorship. What we want to know is what makes the sort of limiting of expression we call “censorship” (often or always) wrong. In order to do this it is not helpful to first define it as wrong at the outset.8 Second, this definition makes the motivation of the censor part of what makes an act count as “censorship.” This is pretty standard in the definitions of censorship; however, it is worth noting a way in which this definition differs from many others that are offered. The above definition specifies that the motivation must be to avoid something “bad,” but does not say it is based on disapproval or moral judgment on the content itself. Many definitions of censorship define censorship as motivated by moral disapproval of the material in question. According to the American Library Association, for example, censorship is “based on a disapproval of the ideas expressed and desire to keep those ideas away from public access” (ALA, Intellectual Freedom and Censorship 8 An alternate approach would be to take my definition and add “wrongfully.” One could, for example, specify that censorship only occurs when a person has a right to expression or when the harms of such restriction in fact outweigh the harm of access. 7 Q & A) [emphasis added]. Similarly, Marjorie Heins defines censorship as “suppressing or restricting speech based on disapproval of its content” [emphasis added] (2004, 230). And, according to Marjorie Boaz, censors want to prevent access to content that is “dangerous to government or harmful to public morality” [emphasis added] (2003, p. 469). While often the motivation for censorship is disapproval of the content or worry about its effects on “public morality,” this is not always the case. I may not morally disapprove of some content, but still think it would be bad if people had access to it. For example, I do not morally disapprove of the information about how to make a bomb out of household chemicals. I do not think having such information available will hurt “public morality.” The information itself is interesting and perfectly benign. Nevertheless, I may think that it is appropriate to limit access to such material. In such a case, I do not “morally disapprove” of the content itself; I morally disapprove of what persons might do with such information. And, this disapproval would not be some particular perspective of mine. It is likely that almost everyone would agree that it would be a bad thing if someone used this information to blow up a power plant. Thus, I think this definition captures the broader range of motivations that might underlie attempts to censor some expression. Third, this definition limits censorship to those efforts to restrict access to an expression “made public by its author.” I have included this proviso in the definition in order to exclude cases where access to private or secret expressions is restricted. To make an expression public is to communicate it with the intent that anyone may access 8 it.9 Above I described censorship as interfering with an act of communication between consenting persons. If I have not willingly made some information public, then you are not censoring, when you prohibit others from accessing it. Thus, for example, if my doctor limits your access to my private medical records, even if he does so specifically because he thinks it would be a bad thing if you accessed the content contained in that record, he is not engaging in censorship. Nor does the government censor when it does not give anyone who asks access to information about its current investigations into a drug smuggling ring. Some may object that, by adding the “publicity” requirement, I have rendered the definition too narrow, because it would not include things that are inappropriately kept from the public. Indeed, there may be cases where someone has the right to know details of my medical history or where the public has a right to know information that the government would prefer to keep secret. Failing to tell people what they have the right to know is different from censorship, however. My sexual partner may have the right to know my HIV status, but to fail to tell him my HIV status or to lie about it is not to engage in censorship. To censor is to be a third party interfering in a communicative exchange. A censor keeps one party from telling another party something. The censor does this either by shutting up the speaker in some way or by interfering with the recipient’s access to that speech. If the speaker does not wish to speak to the recipient or the potential recipient does not wish to receive that speech, there is no censorship. This does not mean that there are not issues of intellectual freedom that concern 9 This would be a case of a completely public expression. Expressions may also be public relative to a particular audience. For example, I may wish to communicate something only to my students or to my friends, in which case it would have been made public to them only. 9 secret or private works, but these issues are about whether such works should be made public. Refusing to make some information public may be detrimental to the deliberative, informational, and expressive interests discussed by Cohen. And, thus, respecting intellectual freedom may in some cases require that such information is made public. This means that censorship does not capture all types of inappropriate limiting of people’s access to information. Rather than being overly narrow, however, this definition allows us to make important distinctions between different ways in which one’s intellectual freedom and right to access may be abridged. Some might also object that this definition is too narrow, because it is overly focused on restricting access, rather than speech. And, thus, it does not clearly classify cases where a speaker is prevented from even speaking—for example, by fear of punishment or by limitation of opportunities—as cases of censorship. As I argued in the previous section, there are good reasons to focus on access to expressions. As Brennen (1965) said, “The dissemination of ideas can accomplish nothing if otherwise willing addressees are not free to receive and consider them” (308). However, while this definition focuses on the restriction of a hearer’s access to some expression, it would still classify punishing or in other ways restricting the speaker as censorship. If the goal of such restrictions or punishments is to prevent others from accessing the potential speech, then it satisfies the above definition. In such a case, access to an expression is restricted by muzzling the speaker. The more serious objection to this definition, however, is that it is overly broad. Many discussions of censorship focus exclusively on actions by the government. James LaRue, for example, defines censorship as “the action by government officials to prohibit 10 or suppress publications or services on the basis of their content” (2007, 3). Some would object that by not limiting censorship to governmental bans on expression, I have overexpanded the concept of censorship. In the rest of this section, I argue to the contrary; this government-focused conception of censorship is arbitrarily narrow. It is true that what we may first think of when we hear the word “censorship” is government putting in place a law that bans an expression from being published or sold. And, it is also true that the state’s “monopoly on power” makes it a particularly effective censor. If the government forbids the public display of a work, its effort at censorship will likely be very effective. By contrast, if I as a private individual refuse to have some work in my house because I don’t want my visitors reading it, I would be a very ineffective censor. However, while the state may have maximal power within a society, such maximal power is not necessary for successful suppression of expression. Suppose, for example, that some entity has control over an entire mode of communication and can restrict certain categories of expression from being included in the media under its control. I fail to see what difference it would make whether it was the government or some other entity exercising this power. It most certainly would not make a difference to the ability of the public to satisfy their interests in access to expression. For the same reason it is not clear that in order to censor one must absolutely prohibit persons from accessing speech by banning it. There are many ways in which one could significantly limit a person’s access to some expression without prohibiting it entirely. One could make the work prohibitively expensive. One could make it socially humiliating to access the expression. One could make it a laborious and time-consuming effort to access it. One could make it very difficult to access the expression without 11 particular skills or technological devices. Furthermore, using all of the above methods, one could limit access without punishing transgressors. True, in order to censor, one needs to have some power, but this does not need to be the power to punish. If one has the financial wherewithal to control some access point to information, one would likely have the power to do at least some of the things listed above. And, any of these acts would sufficiently inhibit the ability of willing speakers to get to willing hearers that it should count as censorship. III. Types of Harm and Arguments Against Censorship I claim that someone censors when they think that it will be a bad thing if other persons access certain expressions. The idea of access to some content being “bad” is very open to interpretation—what sorts of badness do I have in mind? It is important to distinguish two ways in which one might think that access to some content is a bad thing. One may think it is inherently bad for persons to access an expression or that it is merely instrumentally bad for persons to access an expression. In what follows, I discuss how access to content might be thought to be bad in each of these ways and I canvass the arguments that censorship is not justified on these grounds. A. Inherently Harmful Access Often persons would like to censor content that they find offensive and objectionable. They are not primarily concerned to stop the possible bad consequences that might result from such access; they are concerned to stop the access itself. This view holds that accessing some content is simply inherently bad. Examples of content that 12 people have held it would be inherently bad for people to access are explicit sexual content, racist language, violent content, blasphemous works, treasonous works, politically “subversive” works, etc. People frequently find such content offensive on its face. Part of the harm of someone accessing such expressions is thought to be in simply the act of reading or hearing itself, independently of any further consequences that may arise from such access. There are three notable ways in which one might think that access to information is inherently bad: (1) the material is offensive or insulting to the recipient, (2) the material is degrading or corrupting to the character of the recipient (and perhaps the speaker as well), or (3) accessing the material exploits of the human beings who are the subjects of the expression. Given that I characterize censorship as interfering with the communicative acts of consenting adults, I am going to put aside cases of type (1), where persons are the targets of communications that they find offensive or insulting and reasonably do not wish to receive. In particular, I will not deal here with cases such as Stanford University’s policy forbidding hate speech that is, “addressed directly to the individual or individuals whom it insults or stigmatizes.”10 Even most liberal theorists, who tend to promote wide access to information, make a distinction between choosing to access some content and “inadvertent” exposure to it.11 I focus here on cases where some third party wishes to prevent some act of communication between others. So, I will only consider cases where, for example, some racist expression is offered to those wishing to listen to it.12 In such cases, the person who wishes to engage in the censorship is not directly offended or 10 Cited in Brink 2001, 134. See e.g., Brink 2001, Dworkin 1981, Feinberg 1985, and Wendell 1983. 12 The question of whether speakers may have a right in some cases to speak to unwilling hearers is an important issue that I do not explore here. 11 13 insulted by the speech she accesses, but thinks it is a bad thing that other people are able to access such content. If one objects to others accessing “offensive” material, it may be because one thinks that it is damaging or corrupting to a person’s character to access such material. For example, some believe that it is degrading or sinful to look at certain images—an argument often made with regard to viewing pornography. However, it is not the case that inherent badness of access is always merely “self-concerning” in this way. One might also argue that accessing some material is inherently bad with regard to others. In particular, it may be argued that by merely reading or looking at some material we are failing to treat others with respect. Suppose, for instance, some content is created via the exploitation or victimization of persons. If others then access this content in order to get some benefit from it, it may be argued that these persons are participating in and continuing this exploitation. This argument has been made with regard to using or publishing data gathered by Nazi experiments on Jewish people.13 To use such material may be seen as contrary to the Kantian (1998 [1785]) principle of respect; it treats the humanity of the victims of the Nazis as mere a “means” to get information. According to a number of liberal theorists, the fact that some people believe that it is degrading or offensive to access some content does not show that such content should be unavailable to those who do not share this view. Most commonly Mill’s “harm principle” is cited in support of the claim that such restrictions would be unjustifiable. According to Mill, “The only principle for which power can be rightfully exercised over 13 See e.g., Baruch Cohen’s (1990) description of Howard Spiro’s view that, by using the data from experiments carried out on Jews, “we make the Nazis our retroactive partners in the victims' torture and death” (126). 14 any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant” (1975 [1859], 15). Similarly Ronald Dworkin has suggested that the principle of "moral independence" gives us a reason to reject such censorship. On this principle, human beings "have the right not to suffer disadvantage in the distribution of social goods and opportunities…just on the ground that their officials or fellow-citizens think that their opinions about the right way for them to lead their own lives are ignoble or wrong" (Dworkin 1981, 194). On this view, the fact that some persons do not like the idea of other people accessing this content is a very attenuated sort of “harm” that would be unlikely to be enough to justify restricting the access of those who do wish to access it. However, the harm argument does not deal with the sorts of cases I have called “participation in exploitation.” In these sorts of cases the badness is not merely selfregarding—the concern is not merely that I am degrading my character. The idea here is that human beings deserve to be treated in certain ways, independently of whether they are causally affected by that treatment. Perhaps a thought-experiment will help to illustrate what I mean. Suppose we are in a world where there are no children and no more children will be born. (As was depicted in the film “Children of Men.”) Thus, there is no possibility of any child being abused in the future. Would there be any otherregarding badness in looking at child pornography? Clearly my looking at or even buying such material can neither lead to a market in it nor to anyone actually abusing children. However, one might argue that by looking at this material I am failing to show respect to the children who were photographed—I am exploiting them. Some may object to the idea of this rather ethereal sort of harm to others. Nevertheless, it is worth noting 15 that there may be other-regarding concerns about access to expressions, even if one does not think that such access will causally produce harm. B. Instrumentally Harmful Access Not all persons who wish to censor some material wish to do so because they think such access itself is bad. Rather they may be concerned about the bad consequences that might result from such access, e.g., that it will cause harm to someone else. In short, one might think that it is instrumentally bad for persons to access the content in certain expressions. There are a number of ways in which one may think access to some content can result in bad consequences. Below I categorize four sorts of bad consequences one might want to avoid by censoring: (1) creating a market, (2) creating a hostile atmosphere, (3) influence, and (4) implementation.14 1. Creating a Market: Given that persons may be harmed in the creation of certain sorts of content, to provide access to such content is to create a market that will lead to the creation of more content, and thus to more harm to those used in its creation. Most notably this argument has been made in relation to child pornography.15 2. Hostile Atmosphere: The accessibility of certain sorts of content may create an attitudinal environment that undermines the equality and agency of some person or group of persons. For example, some argue that pornography should be censored, because it creates a social atmosphere that perpetuates sexism and the objectification of women.16 14 Note that these are not mutually exclusive; one may think that access causes all of these harms at once. Nor are inherent and instrumental harms mutually exclusive—one may think that access to some content will be bad in all these ways. 15 As was held by the Supreme Court in New York v. Ferber, “the advertising and selling of child pornography provide an economic motive for, and are thus an integral part of, the production of such materials….” A similar argument has also been made with regard to publishing the results of unethical research, see e.g., Florencia Luna 1997. 16 See e.g., Langton 1990 “pornography is seen as a practice that contributes to the subordinate status of women.” 16 3. Influence: The exposure to certain sorts of content may tend to create harmful or antisocial attitudes and behaviors. For example, the most frequent argument for restricting the amount of violence in the media is that it influences children to be more violent.17 4. Implementation: Information may provide instructions or information that can be used to do something that causes harm. This argument is used to support the censorship of works that describe, for example, how to create a bomb, how to commit suicide, how to make drugs, etc.18 Is censorship is ever an appropriate in order to avoid these bad consequences? Before moving on to look at answers to this question, it is worth putting aside at this point a particularly popular, but specious, argument against censorship. According to this argument, no matter how bad allowing people access to some expression might be, we should never engage in censorship because it will start us down a slippery slope. The American Library Association voices a common sentiment when it says on its website that, “What censors often don’t consider is that, if they succeed in suppressing the ideas they don’t like today, others may use that precedent to suppress the ideas they do like tomorrow” (ALA, Intellectual Freedom and Censorship Q & A). According to this argument, there is no way to distinguish appropriate from inappropriate censorship. But, what reason is there to think we cannot make the necessary distinctions? One could just as well say about the law that, “What legislators don’t consider is if they succeed in prohibiting actions they don’t like today, others may use that precedent to prohibit actions they do like tomorrow.” As David van Mill (2002) puts it, “we are necessarily on the slope whether we like it or not, and the task is always to decide how far up or down we 17 18 See, e.g., Etzioni 2004. See Heins 2004 for a critique of this view. For a discussion of restricting information about bomb making, see Doyle 2004. 17 choose to go, not whether we should step off altogether.” But, the slippery slope is not the only argument against censorship based on instrumental harm. There are three standard arguments that it is not appropriate to censor expressions in order to avoid possible negative consequences: (a) in fact there are no such negative consequences that can be tied to accessing such expressions, (b) rights to information cannot be overridden based on such consequentialist reasoning, and (c) the harms created by denying access will in almost every case outweigh any harm created by allowing access. (a) is what Cohen (1993) calls the “minimalist” defense of freedom of speech: negative consequences can never be tied to the simple fact that someone has accessed some information. According to the minimalist, the censor fails to recognize the role of people’s agency in accessing and assessing information. People are not passive recipients of information; they bring their own values, knowledge, and perspectives to the information. Simply because someone reads a book that says that the Holocaust did not occur or a book that describes how to build a bomb, it does not follow that they will come to doubt the existence of the holocaust or go out and build the bomb. If I am, for example, accessing child pornography in order to figure out how to prevent it, then the information is not harmful. One might think of the minimalist slogan as a riff on the National Rifle Association’s “Guns don’t kill people” slogan—“Information doesn’t harm people, people harm people.” However, the fact that any information can be harmless or even beneficial in the right hands does not show that general access or access by the wrong persons to such information may not lead to great harm. Only a very strong libertarian argument would hold that it is never appropriate to limit access to those things 18 which have some benign uses, but that also enable a persons to cause grave harm (e.g., guns, explosives, powerful painkillers). (b) is what Cohen calls the “maximalist” defense of freedom of speech. According to maximalism, even if access to information may lead to harm, it ought never be restricted. This trumping is different from the claim (that I will consider below) that as a matter of fact the benefits of access to expression will outweigh its costs. On the maximalist view, access to information is not the sort of thing that can be weighed against other goods. The maximalist typically bases the value of speech on Kantian concepts such as autonomy. Strauss (1991), for example, argues that to restrict access to persuasive speech involves the “denial of autonomy” because it interferes “with a person’s control over her own reasoning process” (354). One limitation of the rational autonomy argument, however, is that it is not clear that it holds for expressions that cannot be understood as rational or persuasive speech.19 A second more serious limitation of the autonomy argument is that it relies on an overly individualistic epistemology, wherein everyone must evaluate information for herself. In order to see this, we need to clarify an ambiguity in how access to information may lead to harm due to implementation. First, consider the typical sort of case we might worry about. Suppose that Joe wants to check out The Anarchists Cookbook, because he wants to build a bomb to blow up his neighbor’s mailbox. If he gets access to this text, then, given that the directions in it actually give him accurate information about how to build a bomb, he is able to effectively carry out his plans. The information, due to its accuracy, allowed the receiver to attain the knowledge necessary to achieve his goals. 19 Scoccia (1996) argues, for example, that the “persuasion principle” articulated by Strauss, does not cover material that engages in non-rational, non-persuasive speech 19 Things may be quite different, however. Suppose that Susan wants to check out a particular popular book about health, because she wants to learn how to control her diabetes. However, suppose that the information in the popular book is inaccurate. In that case, the information, due to its inaccuracy, may lead the receiver into error and make it harder for her to achieve her goals. In the first case, some may want to limit access to the information because they disapprove of Fred’s goal and want to prevent him from getting the knowledge necessary to reach it. In the second case, some may want to limit access to the information because they approve of Susan’s goal and want to prevent her from getting inaccurate information that will inhibit her ability to reach it. It is less clear that the second sort of “censorship” is interfering with Susan’s autonomy. Rather, it seems that we are promoting her autonomy by helping her to achieve her goals by keeping her away from inaccurate information. In Cohen’s terms, by censoring we would we are supporting her “informational interest;” her interest is in accurate information that will help her carry out her plans. According to consequentialist supporters of (c), “restricting speech is more likely to have bad consequences than protecting it” (Brison, 321). John Stuart Mill (1975 [1859]), for example, famously argues that by allowing a free flow of ideas and information we will be more likely to get to the truth and to keep hold of the truth that we have achieved. More recently Tony Doyle (2001) has argued that, in the context of libraries, even though access to expressions may lead to harm, in almost every case the value of access will outweigh any harms created by it. On this view, whether or not to would be a decision to make like any other, based on whether or not it will produce the 20 most good overall. And, such decisions will depend on the facts of the cases in question. The defender of access to expression may worry that such an approach would too easily allow censorship.20 They worry that by allowing a balancing of access against other goods we put intellectual freedom at risk. Diane Woodward (1990), for example, rejects such consequentialist defenses of access, because it does not “defend intellectual freedom in principle.” “The consequentialist defense of intellectual freedom then, depends upon first establishing whether or not people are better off when they are exposed to all intellectual efforts. The typical result is that one starts dividing intellectual efforts into those that are good for people and those that are not” (15).21 However, it will not be quite so easy to show that censorship will lead to less harm overall than allowing access. First, it is worth reminding ourselves of the central interests in expression, deliberation, and information, which can only be satisfied when there is a general free flow of expression. The benefits we receive from having these interests satisfied (and the harms from not having them satisfied) will not be easily overridden. Second, we have to ask ourselves not what in principle it might be good to censor. We have to ask ourselves what in actual practice would be the consequences of having policies in place that restrict access. It is at this point that “slippery-slope” and “chilling effect” arguments might have some force. The slippery slope may be an actual, not just a conceptual possibility, if human beings in fact tend to not be so good at 20 For an argument, contra Doyle 2001, that utilitarianism cannot provide an absolute defense against censorship, see Fallis and Mathiesen 2001. 21 For a critical discussion of the claim that there is a fundamental right to information, see Himma 2004. 21 distinguishing material they personally dislike from that which is harmful.22 And, there may be a genuine chilling effect on expression if people tend to be overly cautious in avoiding social disapprobation. As Doyle argues, “For intellectual freedom to be genuine, people must have the confidence that they will not be harassed for what they publish or seek out” (2001, 69). This is not to say that even taking all of this into account that using consequentialist reasoning will never lead you to think that censoring is the right thing to do; nevertheless, the hurdle is quite a bit higher than one might have thought. Conclusion In this chapter, I have proposed the following definition of censorship: restricting or limiting access to an expression, portion of an expression, or category of expression, which has been made public by its author, based on the belief that it will be a bad thing if people access the content of that expression. A virtue of this definition is that it does not make describing an act as “censorship” a discussion stopper. In other words, this definition allows that there may be cases of censorship that are morally permissible or even obligatory. People on both sides of a debate can agree that some action fulfills the above definition and they can then go on to have the real conversation about whether such an action is justified. This conversation requires that we look carefully at both why access to expression is important and what the harms related to access might be. Then we can think through the justifications for and against censorship in a clear and systematic 22 Doyle (2001) also argues that more harm than good may be produced if we let people censor, because whoever the “censor” is will be tempted to use this power in ways not justified by utilitarian reasoning. 22 way. My hope is that this chapter has gotten us started on this more fruitful approach to the issue of censorship. Ultimately I believe that, given our strong interests in access to expression and the reasonable concerns about human implementation of policies that restrict access, cases of justifiable censorship will likely be relatively rare.23 23 At least with regard to adult’s access to expressions. There are likely more cases where it will be justifiable to restrict children’s access. See the chapter by Mathiesen and Fallis in this volume for a discussion of children’s rights to access information. 23 Works Cited Association, American Library. Intellectual Freedom and Censorship Q & A. 2007 Available from http://www.ala.org/ala/oif/basics/intellectual.htm. Boaz, Martha. 2003. "Censorship." In Encyclopedia of Library and Information Science, ed. M. J. M. Bates, Mary Niles, Drake, Miriam. New York: Marcel Dekker. Brink, David O. 2001. "Millian Principles, Freedom of Expression, and Hate Speech." Legal Theory 7:119-57. Brison, Susan. 1998. "The Autonomy Defense of Free Speech." Ethics 108:312-39. Carson, Thomas. 2006. "The Definition of Lying." Nous 40 (2):284-306. Cohen, Baruch. 1990. "The Ethics of Using Medical Data from Nazi Experiments." Journal of Halacha and Contemporary Society 19:103-26. Cohen, Joshua. 1993. "Freedom of Expression." Philosophy and Public Affairs 22 (3):207-63. Doyle, Tony. 2001. "A utilitarian case for intellectual freedom in libraries." Library Quarterly 71 (1):44-71. ———. 2004. "Should Web Sites for Bomb-Making Be Legal?" Journal of Information Ethics 13 (1):34-7. Dworkin, Ronald. 1981. "Is There a Right to Pornography?" Oxford Journal of Legal Studies 1 (2):177-212. Etzioni, Amitai. 2004. "On Protecting Children from Speech." Chicago-Kent Law Review 79 (3):3-53. Fallis, Don, and Kay Mathiesen. 2001. "Response to Doyle." Library Quarterly 71 (3):437. Feinberg, J. 1985, Offense to Others: The Moral Limits of the Criminal Law. Oxford: Oxford University Press. Griswold v. Connecticut. 381.U.S. 479. (1965). Heins, Marjorie. 2004. "On Protecting Children--From Censorship: A Reply to Amitai Etzioni." Chicago-Kent Law Review 79. Himma, Ken. 2004. "The moral significance of the interest in information: reflections on a fundamental right to information", Journal of Information, Communication, and Ethics in Society, Vol. 2. Kant, Immanuel. 1998 [1785] Groundwork of the Metaphysics of Morals. Translated by M. J. Gregor. New York Cambridge University Press. Lamont v. Postmaster General. 381. U.S. 301. (1965). Langton, Rae. 1990. "Whose Right? Ronald Dworkin, Women, and Pornographers." Philosophy and Public Affairs 19 (4):311-59. LaRue, James. 2007. The New Inquisition: Understanding and Managing Intellectual Freedom Challenges. Westport, CT: Libraries Unlimited. Luna, Florencia. 1997. "Vulnerable Populations and Morally Tainted Experiments." Bioethics 2 (3 and 4):256-64. Mill, J.S. (1859), 1975, On Liberty in Three Essays, Oxford: Oxford University Press. New York v. Ferber. 458. U.S. 747 (1982) 24 Scanlon, Thomas. 1972. "A Theory of Freedom of Expression." Philosophy and Public Affairs 1 (2):204-26. Scoccia, Danny. 1996. "Can Liberals Support a Ban on Violent Pornography?" Ethics 106:776-99. Strauss, David A. 1991. "Persuasion, Autonomy, and Freedom of Expression." Columbia Law Review 91 (2):334-71. van Mill, David, "Freedom of Speech", The Stanford Encyclopedia of Philosophy (Winter 2002 Edition), Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/win2002/entries/freedom-speech/>. Wendell, S., 1983, "Pornography and Freedom of Expression" in Copp, D. and Wendell, S. (eds.), Pornography and Censorship, Buffalo: Prometheus: 167-183. Woodward, Diana. 1990. "A framework for deciding issues in ethics." Library Trends 39 (1 & 2):8-17. 25