The Indefensibility of Political Representation
[This talk was given at the Austrian Scholars Conference on March 13, 2009. It is available as an MP3 audio download.]
The Principles of Legitimate Command
Tom Hanks and Passepartout are the only two human inhabitants of a Pacific island. Neither is aware of the existence of the other until, one fateful day, they meet. What happens next is a matter of some moment. Will they greet each other politely and go about their respective businesses? Will they agree to cooperate for their mutual benefit? Will they fight? Who can tell? However, we can be reasonably sure in supposing that if Hanks were to command Passepartout to "Tote dat barge! Lif' dat bale!" or demand that Passepartout give up his vile habit of drinking coconut juice while eating fish, or insist that Passepartout cooperate with him in his fishing and hunting ventures, or abstain from servile work on a Sunday — in short, if Hanks were in any way to attempt to require Passepartout to obey his commands, Passepartout would, I believe, rightly resent, and probably resist, such injunctions. The same, of course, applies if roles were to be reversed, and Passepartout were to assume the position of would-be commander.
The Hanks-Passepartout scenario can be replicated in any number of literary variations, limited only by the fertility of one's imagination. For example, more people might be added to the island's population and, while this would result in there being more possible relationships, it would not change the nature of those relationships. The essential principles regarding the legitimacy of command can be established by reflection on our insular drama:
Adam may legitimately command Benjamin to refrain from action C if and only if C is a demonstrable initiation of aggression against the person or property of Adam or against the person or property of another innocent human being.
Adam may legitimately command Benjamin to perform action C if and only if C is an element of a freely (noncoercively) arrived-at binding agreement between Adam and Benjamin, and C does not violate condition 1.
In no other case may Adam legitimately command Benjamin.
If, in 1, Benjamin refuses to refrain from the action C, then Adam may use proportionate force to restrain or punish him.
If, in 2, Benjamin refuses to perform action C, Adam may use proportionate force to elicit compensation.
If, in 3, Adam commands Benjamin, Benjamin may refuse to comply with such a command and, where appropriate, may resist that command with proportionate force.
What is true of the one is true of the many so that if no one person has a right so to command me, no two persons acting severally or in concert have that right. They may, of course, combine to use their superior force to coerce me into doing what they require, but that is a matter of might, not right. Whether the number purporting to command me be one, two, seven, 1223, or 10 million, it cannot, except under the conditions sketched above, be a matter of right.
Rulers and Ruled
So, then, let us consider the situation confronting all of us in our daily lives. In every modern state, some group of people — usually a fairly small group of people — purport to have the authority to command the mass of the population to do this or that or to refrain from doing this or that. They do not possess such a right by virtue of some special divine gift, still less by virtue of their manifestly superior intelligence or moral virtue since sad experience shows that our erstwhile leaders, by and large, are no better in general than the rest of us and are often, sadly, much worse. By what right, then, do they claim the authority to command us, to make laws for us that govern many, if not most, of the significant aspects of our lives?
Government (as the systematic exercise of such command is commonly called) requires a justification. This is not to raise the more fundamental anarchic question of whether governance is at all justified — in the context of this paper, we prescind from this question — it is merely to ask why they are entitled to call upon us to pay taxes or serve in the armed forces or to refrain from taking government-nonapproved drugs or driving without a seat belt. It is to ask why some are rulers and others are ruled.
In the not-so-distant past, those who claimed the right to govern others did so because they had, they claimed, a mandate from God (rather like the Blues Brothers but with wider ambitions); or were better than the common run of man by virtue of their outstanding intellects, sterling characters, Nietzschean will, or distinguished family tree; or had more money than the peasantry; or were simply more powerful than most other people. Whatever persuasive character such justifications may have had in the past, they have none now. Divine-rule theories of government are at an all-time low ebb in the intellectual market, aristocratic theories of government command no respect, oligarchic theories even less, and "might is right" theories are now, as they always have been, absolutely bankrupt. In the arena of governmental justification, democracy is the only game in town, for if there is a fundamental article of faith in the contemporary world, it is not that God is dead or that soccer is the beautiful game; it is, rather, that democracy is a good thing. So entrenched, so widespread, so accepted is this belief that to call it into question is to invite bafflement, bewilderment, bemusement and, when it becomes apparent that one is not joking, dismay, disbelief, and derision.
Democracy and Representation
The key to the justification and popular acceptance of democracy is the idea of representation: those who are governed are thought to be governed by those who represent them and thus, it is claimed, in being governed by those who represent them they are, in effect, governing themselves. This gets over the problem of why, in any political structure, some rule and others are ruled. If rulers and ruled are, in effect, one and the same, then the problem of one person or group of people arbitrarily commanding another disappears. The justification of political governance, then, rests upon democracy, and the justification of democracy in turn rests upon representation. If the bough of representation were to break, then down would come the cradle of democracy, baby and all.
Somewhat less metaphorically, if representation cannot be satisfactorily explicated, then representative or indirect democracy, the last remaining contender for the justification of political governance (in the sense of a division of mankind into rulers and ruled) finds itself in no more tenable a position than any of its discredited competitors.
Despite the central importance of the concept of representation, not a huge amount of work appears to have been done on it. The classic work in this area is Hanna Pitkin's The Concept of Representation, now over 40 years old. She supports my claim regarding the linkage of democracy and representation, noting, "the contemporary popularity of the concept [of representation] depends much upon its having become linked with the idea of democracy" (p. 2). although, as she points out correctly, "[i]nitially, neither the concept nor the institutions to which it was applied were linked with elections or democracy" (p. 3). The contingent connection of democracy with representation is now of historical interest only. For the contemporary mind, democracy and representation are so interlinked as almost to be conceptually indistinguishable.
Given the contemporary firm linkage between democracy and representation, a problem in political philosophy is how best to conceive of political representation. Is a political representative an agent of those whom he represents, limited to the carrying out of their instructions? Or is he a trustee, free to act in the interests of those whom he represents according to his own best judgment of what those interests are? Or is he neither an agent nor a delegate, being simply able to do more or less whatever he likes once elected? Or are there other possibilities in addition to these? Pitkin's book is an extended analysis of the various options.
I believe that the idea of political representation derives such rhetorical force as it has from a set of loose analogies with nonproblematic, ordinary instances of representation, some of which I will sketch below; that none of the ordinary instances of representation translate without loss into the political realm, and that ultimately there is no coherent idea of political representation that can survive rational scrutiny.
Pitkin alleges that, in the 20th century, there was a tendency to
disparage the representativeness of so-called indirect democracies as mythical or illusory. Writers … argue that no government really represents, that a truly representative government does not exist. (p. 4)
I have not been able to find much evidence of such disparagement apart from the anarchic strain of libertarianism but, such as it is, I am happy to add my little contribution to it.
What Is It to Represent?
Are there any constraints on representation? One could envisage a man standing up at a shareholders' meeting and saying "I represent the small investor and I believe that the entire board of directors should be removed" or, in a University, saying, "I represent the administrative staff of the university and we want parity of treatment with the academic staff." One may question whether or not such alleged representatives are in fact representative, but their claim to be representatives of their constituencies seems in principle comprehensible even if it turns out to be false. However, what would one make of a man standing up to say "I represent myself and I believe the entire board of directors should be removed" or "I represent myself and I demand parity of treatment with the academic staff of the university." It would, I suggest, seem a trifle odd.
Of course, one can imagine that in circumstances where it is customary or conventional for one to be represented by another (for example, as a defendant in a legal trial), one might answer the question "Who represents you?" by saying "I am representing myself, my Lord" — clearly, however, this is to be understood as equivalent to the perfectly sensible denial that anyone else is representing me rather than the dubiously meaningful claim that I am, in fact, representing myself. It would seem, then, that a minimal constraint on representation is that there should be a real distinction between the one doing the representation and the one being represented.
Ruling out the relevance of self-representation, let us test our intuitions by examining some ordinary instances of representation:
I am unable to go to a meeting of the committee of my local residents' association. It is an important meeting where decisions of some importance are going to be made so I ask my wife to go along in attendance, subject to the consent of the meeting. I inform her of my views of the important matter under discussion and when it comes up she puts these views forward as being mine. In these circumstances, she represents me.
An issue is coming up for decision in the higher echelons of the university. A discussion takes place at a meeting of the department of philosophy and a general consensus emerges. The chair of the department is mandated to make the department's collective view known to the powers that be. In these circumstances, the department chair represents the department.
I want to buy something at an auction but I do not want to appear there myself for fear of raising the price. I hire a destitute and needy graduate student to purchase a painting for me. I give him explicit instructions on the price. He does exactly what I have commissioned him to do. He represents me for this specific transaction.
I grant power of attorney to my lawyers with general but not completely elastic instructions. As long as they remain within the remit of those instructions, they represent me.
Johnson is my local member of parliament. I did not vote for him. I do not agree with any of his views. Does he represent me?
Robinson is my local member of parliament. I did vote for him, not because I actively desired his election but because I wanted to prevent the election of an even more disagreeable candidate. As it happens, I agree with some but not all of his views. Does he represent me at all times, or only when his actions conform to my views?
In What Way Are Political Representatives Supposed to Be Representative?
In what way are our political representatives representative? What does it mean for one man to represent another? Under normal circumstances, as our examples show, those who represent us do so at our bidding and cease to do so at our bidding. They act on our instructions within the boundaries of a certain remit and we are responsible for what they do as our agents. Furthermore, the central characteristic of representation by agency is that the agent is responsible to his principal and is bound to act in the principal's interest. Is this the situation with my so-called political representatives? Political representatives are not (usually) legally answerable to those whom they allegedly represent. In fact, in modern democratic states, the majority of a representative's putative principals are in fact unknown to him. Can a political representative be the agent of a multitude? This also seems unlikely. What if the principals have interests that diverge from each other? A political representative must then of necessity cease to represent one or more of his principals. The best that can be done in these circumstances is for the politician to serve the many and betray the few.
A political representative — at least the typical member of an elected legislature — has a constituency rather than a single principal; and that raises problems about whether such an unorganized group can even have an interest for him to pursue, let alone a will to which he could be responsive, or an opinion before which he could attempt to justify what he has done.… the political representative has a constituency, not a principal. He is chosen by a great number of people; and, while it may be difficult to determine the interest or wishes of a single individual, it is infinitely more difficult to do so for a constituency of thousands. On many issues a constituency may not have any interest, or its members may have several conflicting interests. (pp. 215; 219–20)
In Pitkin's view, these passages establish the difficulty of representing a constituency. However, she understates the problem. It is not that it is difficult to represent a constituency — it is rather that it is impossible, and she herself has given the clue why this is so. There is no interest common to the constituency as a whole, or, if there is, it is so rare as to be practically nonexistent. That being the case, there is nothing to represent.
Some may take issue with the notion of representation presented here and argue that we are dealing with a considerably more complex phenomenon, that political representation is just one instance of a variety of types of representation, that representation can be symbolic, formal, religious, or iconic. Firstly, while my remarks apply primarily to representation-as-agency, similar considerations can be brought to bear, mutatis mutandis, on representation as trustee, deputy or commissioner, and so on. Once again, as with our desert-island drama, the basic conceptual point can be grasped from the single example of representation-as-agency — there is little to be gained, except a soothing tedium, from a rehearsal of the inapplicability of the other paradigmatic types to political representation. Secondly, one could agree that there are currently a variety of notions of representation. I have mentioned symbolic, formal, religious, and iconic as types of representation. A full treatment would require a discussion of all these, and other, types of representation. Space does not permit me to do this here, but I would like to make a few remarks about just one of these types currently enjoying a wave of popularity, namely iconic representation.
In iconic representation, A is said to represent B if A is like B in some particular respect; so, a woman, simply by virtue of being a woman, represents other women; a person of a particular skin color, simply by virtue of that fact, represents other people with the same skin color. But there is a logical problem here. Everything is like everything else in some respect or other, and so it comes about that, on this notion of representation, anything or anybody represents any other thing or anybody else. Such a notion of representation evacuates it of any real significance. What sense can be made of claims sometimes made that some group, say women, are "underrepresented" in particular professions? In most contexts, there is simply no representation at all going on. Suppose that I, a man, am employed in a particular capacity in a particular firm — just by virtue of being a man I do not represent men. By the same token, I don't represent fathers, philosophers, the middle-aged, the cranky, or any other group.
These are not appropriate arenas for representation and so there can be no under-representation simply because there can be no representation. (Oddly enough, one rarely hears complaints of groups being underrepresented in such nonglamorous occupations as trash collection or sewage works.)
Other kinds of representation — religious, symbolic, etc — may well play a role in human discourse and action but this sheds no particular light on the central problem we are presently concerned with, which is that of political representation. I cannot imagine anyone being satisfied with an account of political representation that ultimately reduces it to the symbolic, the religious, or the iconic.
It is, of course, perfectly possible that the concept of representation is systematically ambiguous and that there is at best a sort of family resemblance between its various kinds. If this were so, it would leave the notion of political representation as a more or less distant cousin of other kinds of representation so that, as in the case of human relations, while John resembles Howard, and Howard resembles Tim, and Tim resembles Michael, it doesn't follow that John resembles Michael in any way. However, Pitkin adopts as a working assumption the position that
representation does have an identifiable meaning, applied in different but controlled and discoverable ways in different contexts. It is not vague and shifting, but a single, highly complex concept that has not changed much in its basic meaning since the seventeenth century. (p. 8)
Her attempt at a definition is as follows: "representation, taken generally, means the making present in some sense of something which is nevertheless not present literally or in fact" (pp. 8–9). This is immediately followed by another attempt at definition that may or may not be the same: "in representation something not literally present is considered as present in a non-literal sense" (p. 9). Pitkin admits that this/these simple definition/s may not be particularly helpful. It is hard to disagree with that negative assessment.
Having exhaustively examined the various instances of unproblematic representation — agent, trustee, deputy, commissioner, and so on — Pitkin concludes,
None of the analogies of acting for others on the individual level seems satisfactory for explaining the relationship between a political representative and his constituents. He is neither agent nor trustee nor deputy nor commissioner; he acts for a group of people without a single interest, most of who seem incapable of forming an explicit will on political questions. (p. 221)
It is difficult to see how this point could be made more clearly. One would think that such a state of conceptual confusion would lead one to give up any idea of discovering a coherent account of political representation. But Pitkin ploughs on:
Must we then abandon the idea of political representation in its most common sense of "acting for"? This possibility has sometimes been suggested; perhaps representation in politics is only a fiction, a myth forming part of the folklore of our society. Or perhaps representation must be redefined to fit our politics; perhaps we must simply accept the fact that what we have been calling representative government is in reality just party competition for office. (p. 221)
One is tempted to say, Yes! Yes! Alas, Pitkin says, No! No! She thinks that it is, perhaps, "a mistake to approach political representation too directly from the various individual-representation analogies — agent and trustee and deputy" (p. 221).
She then proceeds to suggest a kind of institutional or systemic account:
Political representation is primarily a public, institutionalized arrangement involving many people and groups, and operating in the complex ways of large-scale social arrangements. What makes it representation is not any single action by any one participant, but the over-all structure and functioning of the system, the patterns emerging from the multiple activities of many people. It is representation if the people (or a constituency) are present in governmental action, even though they do not literally act for themselves. (pp. 221–22)
She picks up this idea again when she says,
[W]hen we speak of political representation, we are almost always speaking of individuals acting in an institutionalized representative system, and it is against the background of that system as a whole that their actions constitute representation, if they do. (p. 225)
Frankly, this is nonsense and, ultimately, a counsel of despair. It comes to this. None of the paradigmatic uses of the term "representation," as instanced by the various examples Pitkin considers (deputy, agent, etc.) suffices to make sense of the idea of political representation. So, Pitkin invents a whole new unsubstantiated systemic account. Instead of individuals representing, we have a system that represents. We are to forget that we have been unable to make any sense of individual political representation; we can kick the problem upstairs by ignoring the individual and having the system itself be representative. Let us risk committing the fallacy of composition and assert that if the idea of explicating political representation by means of the analysis of individual acts of agency, trusteeship, and so on is unrealizable, the problem is hardly solved by simply positing "the system" as the superagent of representation.
I would go further: the systemic account is not only unhelpful; it is obfuscatory, appearing to explain when in fact it simply sweeps the problem under a pseudo-explanatory carpet, in a manner reminiscent of the postulation of "dormitive power" by the doctor in Molière's Le Malade imaginaire as an explanation of the soporific qualities of opium. This, of course, is to explain the obscure by the more obscure; it is also a striking example of what Alfred North Whitehead called "the fallacy of misplaced concreteness."
If it is to be tenable, representative or indirect democracy requires a clear, robust, and defensible conception of representation. No such conception has been forthcoming, and it is doubtful if any ever will be forthcoming. It used to be said that only three things were definitely true of the Holy Roman Empire: it wasn't holy, it wasn't Roman, and it wasn't an empire. Similarly, two things are definitely true of representative democracy: it isn't democracy and it isn't representative.
In the end, representation is a fig leaf that is insufficient to cover the naked and brutal fact that even in our sophisticated modern states, however elegant the rhetoric and however persuasive the propaganda, some rule and others are ruled. The only question is, as Humpty-Dumpty noted in Through the Looking-Glass, "which is to be master — that's all."
 This is a version of the basic axiom of libertarianism. Almost any treatise on the subject contains an exposition and defense of this axiom. See, for example, Boaz, D. (1997). Libertarianism: A Primer. New York: The Free Press; Rothbard, M. N. (1982). The Ethics of Liberty. Atlantic Highlands, N.J.: Humanities Press. [Reprinted (1998). New York: New York University Press.]; Rothbard, M. N. (2004). Man, Economy and State. [2nd edition (Scholar's Edition)] Auburn Alabama: Ludwig von Mises Institute.
 Pitkin, H.D. (1967). The Concept of Representation. Berkeley and Los Angeles: University of California Press.
 Those who know will, of course, recognize my debt here to the writings of Lysander Spooner.
 An instance of symbolic representation occurs when Elrond is choosing the Company of the Ring in Tolkien's (1969) Lord of the Rings: The Fellowship of the Rings. He says: "For the rest, they shall represent the other Free Peoples of the World: Elves, Dwarves, and Men. Legolas shall be for the Elves; and Gimli, son of Gloin for the Dwarves….For men you shall have Aragorn…" (London: Harper & Collins, p. 362).
 Molière (1673) Le Malade imaginaire — "quiat est in eo virtus dormitiva cujus est natura sensus assoupire." ("Because there is a dormitive virtue in it whose nature is to cause the senses to become drowsy.")