Recently, the British Times launched an attack on the ‘excepted hereditary peers’, those few dozen hereditary lords who have been elected to sit in the House of Lords by their fellow hereditaries, following the hereditaries’ loss of their automatic right to sit in parliament in 1999. The Times described these peers’ presence in parliament as an unfair and absurd anachronism. American readers may well be inclined to agree with the Times on democratic grounds. However, what should interest and concern all those who value freedom, was the flawed theory of political representation implicit in the Times’ case. This theory is becoming ever more prominent in today’s political discourse in the West; it is also extremely dangerous. Today I want to critique it, to offer an alternative theory, and to consider what that alternative implies about the place, within contemporary democratic politics, of supplementary non-democratic institutions like the hereditary peerage.
1. The Identity Theory of Political Representation
The keystone of the Times’ case was its argument about political representation. Beneath a front-page collage of passport-style photographs of the eighty-five sitting hereditaries—a uniformly white, male group of a kind almost never depicted in today’s British media—the Times’ leader claimed that they were ‘unrepresentative’: they ‘do not look like modern Britain’. They are all white and male; most are old; almost half went to Eton.
The Times, in other words, argued that the hereditary peers fail to represent today’s British population just because they are white, old, privately-educated males. Now, it is of course possible that the Times’ editors would want to nuance this bald argument on cross-examination. Nevertheless, the syllogism that underlies the Times’ argument is now standard in political rhetoric: ‘X doesn’t look like Y; no-one can politically represent someone he doesn’t look like; therefore X cannot politically represent Y’. Let us therefore hold the Times strictly to its fashionable syllogism, and explore the reasoning behind it. As we shall see, it is flawed and dangerous.
The reasoning behind the syllogism, if one can glean its fragments from the barren formless fields of progressive thought, is roughly as follows. First: politics aims at justice. Second: there are certain natural categories of man which are such that, if one man differs from another according to any one of them, then he has has a radically different ‘lived experience’ from the other. These natural categories, says the progressive, include race, sexual orientation, gender identity, and perhaps several others. Third: if X differs from Y in a natural category, and therefore has a radically different ‘lived experience’ from Y, then it is impossible for X to understand why Y takes the views he does about political justice: whether certain laws and policies are or are not just. Therefore X cannot understand Y well enough to represent Y in politics1.
Let us formalize this claim, calling it the identity theory of political representation. According to this theory, one person represents another in politics only if he belongs to same key categories as him. These key categories include race, gender identity, sexual orientation, and several others; let us call the associated properties, such as blackness or manhood, key properties. A necessary condition of representation, then, is identity of key properties, or key identity. Variants of the basic theory may define other conditions of representation, but key identity is always the first and most important one in any identity theory.
The theory describes not only what it is for one individual to represent another, but also what it it for a legislature or elite to be representative of its people. A legislature or elite is wholly representative of the people only if it contains the same mixture of key properties as the people. All other things being equal, the greater the difference of mixtures of key properties between the legislature/elite and the people, the less representative the former is of the latter.
Typically, one encounters the identity theory of representation alongside the representation theory of legitimacy and justice. This theory states that a legislature or political elite holds its power legitimately and justly only if it is representative.
2. The problem with key categories
The identity theory of representation, taken alongside the representation theory of legitimacy and justice, is problematic in various ways. First, the question obviously arises: Which human categories are key categories? Indeed, if our two theories are correct, then answering this question must surely be one of the first tasks of politics.
Unfortunately, the theories imply that it is impossible for this question to be resolved by rational political argument within the public sphere. To see this, suppose that the members of some identity group (LGBTQ, transgender, black, etc.) argue that they feel in some way different or excluded from others, who cannot properly understand them—suppose, in other words, that they argue that they have a radically distinctive ‘lived experience’. This, in effect, is to argue that the human category under which they count as minorities—perhaps sexuality or gender identity or race—is a key category.
Within today’s Western politics, an argument of this kind would not be not a formal argument, which achieves some codified result if some legally established decision-making body is convinced of its merits; that is, it would not be like a motion of no-confidence, or a parliamentary recall motion. As yet, no country has any legal procedure by which it might be established that some category is a key category. There are no legislative votes about whether a category is ‘key’; there is no way to enshrine a key category, as such, in law. Rather, an argument of this kind would be an informal attempt to establish a category as key within political discourse—to convince the powers that be, the media, ‘opinion-formers’, and ultimately the public, that a certain category is key. We can call such an attempt, an attempt to vindicate a key category.
So then: our supposition is that some identity group is attempting to vindicate a key category, in order that its members might gain a claim to representation as such. Suppose, further, that this attempt has met with no success; suppose that the political elite—formed of politicians, media leaders, and so on—has concluded that the group’s lived experience does not merit separate representation; that the elite has not shown any concern to take account of the lived experience of a certain group, or to share the group members’ typical political views and attitudes. In other words, the elite has rejected the attempted vindication of the key category on rational grounds.
Here, then, lies the problem. How does one know that the elite’s rejection of an attempted vindication of a key category, does not merely reflect the elite’s failure to understand the lived experience of the group? How does one rule out the possibility that the elite simply cannot understand the group’s lived experience, because it really is unique? For perhaps the category under which the group is a minority, really is a key category. And if this is so, and if the elite contains disproportionately few members of the group, then the elite is therefore unrepresentative, and to that extent illegitimate.
In other words, any identity group of which the elite has disproportionately few members, will have grounds to reject the elite as illegitimate, if that elite rejects the group’s attempted vindication of its key category; for this rejection will only prove that the category really is a key category, and that the elite, which has disproportionately few members of the identity group, therefore cannot understand the group’s views and attitudes.
As this demonstrates, politicians, and all who wield power, cannot fully determine which categories are key by means of a rational debate in which both sides of the argument have reason to engage. For, whoever the elite are, inevitably some natural (or purportedly natural) properties will be shared by disproportionately few of them. Hence there will always be categories the attempted vindication of which cannot be resolved by a debate in which both sides accept the authority of shared, objective forms of reasoning. For if the vindication of such a category fails, then the would-be vindicators of it will have grounds to label those who set public discourse as holders of illegitimate power, who cannot possibly understand them and their ‘lived experience’. The two theories imply that there are invisible walls of logic between members of groups defined by different key properties; walls through which mutually-accepted political reasoning cannot penetrate.
If our theories accurately describe the progressive view of representation, then, and if this view now has widespread currency in Western discourse—which I claim is true—then we would expect to see two things happening in Western politics. First, we would expect many groups to be attempting to vindicate the key categories under which their members count as minorities. For by doing so, such groups would be able to claim that their members are unrepresented, that they are therefore victims of an illegitimate system, and that they are therefore entitled to some political redress. In other words, the vindication of key categories would give groups power. Second, we would expect that the struggle over which categories are key, would not be conducted through rational argument.
I suggest that both these predictions hold good. In today’s public discourse, there is a constant clamour for attention of the part of various purportedly marginalized groups. Which gain the most attention, seems largely to depend on the prejudices and interests of those who set public discourse, and on which groups shout the loudest; rational debate does not seem to come into the matter. Indeed, such irrational power-struggles are becoming an ever more prominent feature of Western public life.
3. The Problem of subtractive intersectionality
The second problem with our two theories aggravates the first. As the list of key categories grows (which it will do, because identity groups have a great incentive to vindicate them), then it becomes impossible for all key identities to be represented in the legislature or elite. For as we have seen, a person cannot represent another, according to the identity theory, if he differs from him in any one key property. Indeed, this is the only viable fleshing-out of identity theory: it would be obviously arbitrary to say that one person can represent another provided that he differs in, say, no more than two or three key properties. Identity theory thus endorses subtractive intersectionality. According to subtractive intersectionality, the ‘lived experience’ of a person who is a minority under multiple key properties—such as a black, ‘trans’ person—is radically different from the ‘lived experience’ of those who lack his exact combination of minority statuses. Hence anyone can assign himself a recherché, compound key identity by combining all of the key properties under which he is a minority. Perhaps—he might say—my black M.P. understands my blackness, but he isn’t a black disabled trans-woman. How can he possibly understand me? Indeed, this manoeuvre is so easy, and so intellectually cheap (because rationally irrefutable), that it will surely become constituents’ default response whenever they disagree with their political representatives. The identity theory thus encourages people to feel unrepresented, and hence the victim of an injustice. It implies that it is impossible for a legislature or elite to represent more than fraction of a populace (for there are a very large number of possible combinations of the ever-growing key properties). Therefore, taken with the representation theory of legitimacy, it ultimately implies that political power within representative democracies never has more than a minimal legitimacy.
4. The Problem of Political Compromise
The third problem with our theories is the most serious. Here we will talk only about specifically legislativerepresentation within a representative democracy—unlike in the above sections, we will not be interested in political ‘representation’ in a broader sense.
As we have seen, the identity theory of representation relies on the premise that no man can well represent those who have a different ‘lived experience’ from him. But this premise itself is disastrous. We can see this if we consider the purpose of legislative representation, and, more fundamentally, the reason why we have politicians. First, then: To what end do we employ politicians? Answer: that we might be governed justly. The ancient and ecclesial tradition from Plato onwards tends to see justice both as as a virtue of persons, and as a property of states and their laws. Taken overall, the tradition assumes that justice is objective, but not necessarily determinate: the question ‘Is this something that a just man would do?’ or ‘Is this a just law?’ has (or usually has) an answer; but in any one case there may multiple possible just actions or laws. Rulers needs intelligence and wisdom to govern justly, and so to devise reasonably just laws and policies. Fortunately, however, most people can recognize just laws even if they cannot devise them. Hence assemblies and senates and counsels and parliaments arise, to debate difficult questions, and to provide solutions that the people can see are reasonably just.
Such bodies also arise for a related reason. Any society presupposes much agreement on fundamentals: citizens must agree that laws should be obeyed, for example. But men have always understood that views differ on political questions. Parties form, grouping together men of similar views—men who tend to come to similar conclusions about what is just in a range of important matters. If a party is powerful, if it believes itself to have a right to be heard, and if its views are consistently ignored, there is a danger of civil war. Hence in all stable societies (other than ones whose people are deeply deferential to some ruler), powerful parties are in some way represented amongst the political elite, often by means of a parliament. The political representatives of the various parties or factions find compromises that make common life tolerable. Politics is essentially eirenic; it is the art of compromise, and of wisely balancing interests. The politician who always strives to undermine his opponents’ aims by any means possible, heedless of any resentment he creates amongst powerful opponents, is a bad politician, because he undermines the political system himself. Overall, then, a good politician strives for acceptable compromises that are as just as possible, according to his conception of justice. (Of course, a good politician with a strong majority behind him may well push through his legislative programme; but he will still need to compromise on, for example, procedural matters, and probably also on small points of detail in his legislation. He would not be a good politician if were intransigent).
Here, though, we should note something about compromise. There is no algorithm for the creation of political compromises. Politicians cannot simply feed opposing policies into some political machine, and await a printed-out compromise. On the contrary, politicians need creativity and imagination if they are to help make good compromises. To compromise, a politician must, to some extent, see a question from his opponents’ points of view. Imagine, for example, a case in which one party wanted cars to drive on the left, and another party wanted cars to drive on the right. A naive algorithm for compromise might tell them that cars should drive in the middle. But human politicians would never propose such an absurd law, because each party would understand that the other would not want such a compromise: each would know that the other would think it unjust.
How is this mutual understanding possible? After all, political opponents are those who tend to come to different conclusions about what is just in a range of cases. But to compromise, one must be able to predict how one’s opponents will react to new proposals: whether they will find them acceptably just. How can one guess what conclusions one’s opponents will come to about any one of the infinite questions a parliament may face? It is clear that politicians domanage to do so in practice—but how? The answer, I think, must be something like this. A good politician predicts his opponents’ conclusions by reasoning wisely and intelligently from his opponents’ politically relevant premises. Thus he works out the outline of policies, laws, etc. that would be just if his opponents’ premises were true. So to speak, he uses his imagination to enter into his opponents’ world of thought.
Let us therefore assume that wise and intelligent politicians can usually achieve political compromises acceptable to the people on most issues; for this seems to be the case, and politics would be futile if it were not. Given the above, this claim then follows: If a wise and intelligent politician A, reasons from the politically relevant premises of a wise and intelligent politician B, then, on almost any practical question of governance, A will come to similar conclusions as B about what action or policies would be just. Indeed, it is at least notionally possible that politician A may even be persuaded of politician B’s views, and change his political allegiance. This holds irrespective of any other differences between A and B. With this claim in place, we can now make two further claims. 1) Conclusions about the just areprimarily premise-determined: one’s practical conclusions about what is just will depend upon one’s politically relevant premises, plus one’s intelligence and wisdom. 2) Conclusions about the just are identity-insensitive: one’s practical conclusions about what is just will not depend upon one’s natural properties as such, other than intelligence—and perhaps wisdom, if that counts as a natural property. Nor will they depend upon any unique ‘lived experiences’ tied to one’s properties. 2) is a corollary of 1).
Here I should note something important about point 2). In a given culture, all the members of some identity group may well tend to share distinctive premises; indeed, we have already implied as much. Point 2) does not deny this; but it claims that others who adopt an identity group’s distinctive premises, will able to follow the group’s reasoning to its conclusions. So, for, example, 2) claims that one does not need to be LGBT to predict the ‘Pride’ lobby’s conclusions on practical questions; one only needs the imaginative intelligence and wisdom to discern its premises and then to work out what would be just if they were true. Unless 1) and 2) are true, then politics is futile, because creative compromise is impossible.
Here, then, is the disastrous problem with the identity theory of representation. Identity theory claims that a representative must be key-identical with those whom he represents. Indeed, this is always the first and most important condition of representation in any version of the theory. According to identity theory, if I am looking for someone who will represent me then my first criterion should be key identity. But why tie representation to identity in this way, unless one believes that every man’s practical conclusions about what is just, depend more on his identity than his politically relevant premises? The identity theory only makes sense on the assumption that one’s identity as such, not one’s set of premises, is somehow the primary determinant of one’s practical conclusions. As we began to see when we considered the first problem with the theory, the theory’s fundamental flaw is that it removes people’s political views from the sphere of the rational and mutable, and puts in them into the sphere of the innate and immutable.
But now the problem is evident! Identity theory makes sense only if claims 1) and 2) are false. But if so, then creative compromise is impossible; and in that case, politics is futile—as, a fortiori, is political representation. The identity theory of political representation is self-defeating.
By saying this, I mean firstly that the theory must be wrong, because it implies a general conclusion that is false—namely, that all politics is futile. But I also mean that the theory, if widely accepted, makes the populace such politics really does become, in one time and culture, futile: for it sows the seed of the idea that, if one feels frustrated with one’s politicians, it is probably because they cannot understand one’s ‘lived experience’, and are therefore illegitimate. It undermines that condition that I said above must obtain if politics is not to be futile: that politicians can come to compromises that the people will accept.
5. The Interest Theory of Representation
Given the above reflections, let me briefly propose a different theory of representation: the interest theory of representation. According to the interest theory of representation, to ‘represent’ a person in public life is not to ‘look like’ him, or to share his key properties; rather, it is to articulate his views and attitudes in the public sphere, and to defend his interests, alongside the national interest; ideally, it is to do so better than the average man could do.
Consider the implications of this theory. For reasons already explained, a politician will need to be able to compromise in order to serve his constituents’ interests and the national interest well. Hence, according to the interest theory, a good representative will be wise and intelligent, so that he will have a fairly good faculty of discerning opponent’s premises, and working out what would follow from them. On this theory, representation is no sense identity: the average uneducated person would do a bad job of interest-representing himself in parliament. So would the average young and inexperienced person. Hence a wise, well-educated old man may be a better representative of the young and uneducated, than the young and uneducated themselves.
Let us assume that this theory is roughly correct. Arguments like the Times’ are then obviously absurd. The mere fact that the hereditary peers do not ‘look like’ modern Britain tells us nothing about whether they are representative of modern British people. The Times might reasonably have argued that in Britain old white men tend to have different views than much of the populace, and that they are bad representatives of the public for that reason; but read strictly, their argument is false.
Here it is notable that common traditional practice seems to endorse the interest theory of representation—or at least something much closer to the interest theory than to the identity theory. Many countries’ upper-houses are called ‘senates’, and ‘senate’ derives from the Latin senex, ‘old man’. Old men obviously differ from most of the population in intuitively-important natural properties; nevertheless, common traditional practice assumes that they will make for good representatives. For—prescinding from the question of a representative’s views—if he has long experience of life and a first-rate education, then that is a good thing; such attributes make for wise, articulate—and therefore good—representatives.
The interest theory of representation also provides an interesting, nuanced perspective on democracy. The Times’argument draws an implicit contrast between the ‘unrepresentative’ hereditaries and Britain’s MPs—i.e., the members of the House of Commons. But are MPs thoroughly representative just because they are elected? In fact, MPs are disproportionately culturally and socially liberal; as are most of today’s elected Western legislatures. According to the interest theory, the hereditaries, who form only a small part of parliament, therefore have a useful representative function: they provide some counterbalancing representation of the remaining patriotic cultural conservatives in Britain. In other words, the theory suggests that, in today’s West at least, there may be good reasons for conceding a modest, supplementary legislative role to some unelected group in a truly representative political system. Arguably, the hereditary peers’ place in the Lords is one institution defensible under the theory.
Where does this leave us? Clearly, far more could be said on all of these matters. My modest aim for today has been to show the flaws and dangers in the ever more prominent ‘identity’ theory of political representation, and briefly to show that the alternative, ‘interest’ theory bids us not rush to condemn every non-democratic institution within representative democracies, nor every institution whose members do not ‘look like’ the populace. Ultimately, the battle over how to understand ‘representation’ is a battle that will determine whether we conservatives uphold the idea of rational political argument itself, or whether progressives lead us into a new dark age of aphasic immutable political tribalism.
1 Logically, the converse ought to be true of Y. However, progressives generally only employ the reasoning sketched above, when a person of privileged natural categories, such as a white man, is a political representative a ‘marginalized’ other, such as a gay black ‘trans-man’; not when the roles are reversed. For progressives, the ‘can represent’ relation seems to be asymmetric—which is why I have made it so above. However, I will not say anything more about this complication today.