The political left is currently engaged in an internal struggle about identity politics. Defining “identity politics” is to some extent subjective – it sets the ground for the subsequent debate on its merits.

It turns out that trying to discern a succinct definition which captures the disagreement about identity politics is hard. I was able to extract the following ideas as elements of identity political belief, as defined by some supporters (here, here and here) :

• Women and certain minorities (ethnic / racial, LGBT, I, Q …) are under-represented at the moment in power structures (parliaments, media, etc), relative to the weight in the population.
• Economic inequality and structural (minority and/or gender based) “oppression” (or obstacles to opportunity, if you prefer a less emotive term) are linked.

I don’t think these statements are really disputed on the left, so they don’t really define the source of disagreement.

However, if one disagrees with the identity politics approach, as I did recently, it is often assumed that one also disagrees with both of the points above. This is therefore the first difficulty in talking about identity politics – if you don’t agree with the proposed solution to the problem, you must be denying the existence of the problem.

A second difficulty – which i’ll discuss below – is that what you say in the discussion is very much secondary to who you are, so it’s hard to focus on ideas rather than the relative social position (or in practice, the “group”) of the person proposing the idea.

Both of these problems often prove insurmountable.

—

Judging from a brief stint of internet research, the term “identity politics” appears to have been coined in 1977 by a black feminist group called the Combahee River Collective who wrote: “This focusing upon our own oppression is embodied in the concept of identity politics.”

An article by Mychal Denzel Smith expanded on this concept. He said identity politics was not intended to be “group specific” because the Combahee River Collective thought that if they could liberate their own identity group, then “ … it would mean that everyone else would have to be free since our freedom would necessitate the destruction of all the systems of oppression”. This was peculiar to their identity as black women – a group at the intersection of two types of oppression (racial and gender).

Smith then goes on to say that identity politics can be applied broadly as

… the place to understand what forms of oppression are operating within your own life. From here, coalitions can be built with others who face similar forms of oppression, so long as it is also understood that oppression is not experienced the same across identities.

In response to these ideas, I had the following thoughts:

• It does not necessarily follow that removing all forms of oppression from the most oppressed group would mean no-one is oppressed. One could of course simply substitute one group for another in the hierarchy. However even if I take their ambition in good faith, the objectives of the Combahee River Collective are clearly anti-capitalist and pro-Marxist, which is their vision of “non-oppression”, not some sort of objective state we could all agree was free from oppression. Economic liberals consider Marxism to be one of the worst forms of oppression (and indeed, that’s how it did turn out when it was practised). The Combahee River Collective vision essentially swaps one concept of socio-economic oppression for another. That’s fine – but it undermines their claim that “everyone else would have to be free”.
• The group identity approach seems to assume groups are monolithic or should be, i.e. there are no sub-groups or hierarchies within groups. The Combahee River Collective statement has some interesting things to say about sub-groups or non-racial group identities, such as black lesbianism. In fact they “… reject the stance of Lesbian separatism because it is not a viable political analysis or strategy for us. It leaves out far too much and far too many people”. What is the “right size” for an identity group then ? (black women aged 15+ being 4.3% of the US population in 1977).
• The idea that one can form coalitions across oppressed groups facing “similar” oppression seems at odds with the sine qua non of identity politics – that only the group itself can understand the oppression experience and represent itself (without inadvertently “working to end someone else’s oppression”, to quote the Combahee River Collective Statement).
• The idea of a “coalition against the oppressors” suggests that it’s clear who they are, and who is being oppressed by whom, that all of this is cut and dried and everyone falls into neat camps (oppressed versus oppressor). I would have thought that logically (and practically) one could be both oppressed (by person A) and oppressor (of person B) at the same time.

—

Perhaps the best definition of identity politics that I found from someone sympathetic to the idea was the one in Kimberlé Crenshaw’s famous paper on “intersectionality”. She quickly outlines two different liberal schools of thought:

The embrace of identity politics, however, has been in tension with dominant conceptions of social justice. Race, gender, and other identity categories are most often treated in mainstream liberal discourse as vestiges of bias or domination-that is, as intrinsically negative frameworks in which social power works to exclude or marginalize those who are different. According to this understanding, our liberatory objective should be to empty such categories of any social significance. Yet implicit in certain strands of and racial liberation movements, for example is the view that the power in delineating difference need not be the power of domination; instead be the source of social empowerment and reconstruction.

In this description, I could easily recognise my position (“mainstream”) and that of identity politics. Also, I think this statement points to the heart of the debate – does “delineating difference” serve to undermine structural inequality, or to reinforce it, or does it depend on the user?

My views on this are coloured by my own (limited) experiences as “an outsider”. In particular, an event when I was a student in France and was invited over to dinner at a friend’s place. There was nothing remarkable about the dinner per se, but the event sticks clearly in my mind because it was the first time I had sat down at a social event (in France) with a bunch of French guys, and ate, drank, chatted, and didn’t once have to answer any questions about Australia (where I am from). Now, obviously there is nothing wrong with being asked questions about Australia, but the absence of such questions made me suddenly feel “at home” in France. So I realised that the unconscious subtext to such questions is always “you’re not one of us”. They serve to remind you that you’re not part of “the group”, and group belonging is clearly an innate human desire. This explains the immense pleasure I got out of that evening – it was the first time I simply passed as part of the group, or rather, was not seen as a representative of a different group.

Based on this experience, it has always been my view that a civilised and warm way to treat others is not to delineate differences. And this places me in the mainstream liberal camp referred to by Crenshaw. Apart from that personal experience, here is a list of the broader problems I have with identity politics.

• It denies the ideas of objective (i.e. shared) reality or logic – everything depends on “who you are”. I had a taste of this in a recent discussion, when I was told that my understanding (and indeed the dictionary definition) of the word racism was wrong, because the word was not defined by the “victims”. Firstly, there is a logical problem here: until we know what the word means, we can’t know who is a victim. Secondly, insisting that victims define a word is an (arbitrary) normative choice  – I mean we could also propose that the word racism be defined by racists (after all, it refers to them !). Thirdly, where, if anywhere, does this logic end? Should the word “elephant” be defined by elephants, or by their victims? Do we have go through the dictionary and identify a victim for each word in order to reach a common understanding of what the word means? And what if – God forbid – someone is both a victim of racism, and a racist themselves (a possibility not countenanced by identity politics, which can’t see the complexity that individuals, or even groups, presents). A Hutu engaged in a genocidal (and racist) effort to kill Tutsis could simultaneously be the victim of European white racism towards Africans generally.
• Identity politics then goes further – apart from establishing an exclusive right to define the language we all use to those deemed to be victims – anyone who disputes this right is committing violence. Not “violence”, actual violence. This marks a major redefinition of the word violence, which has important political consequences. One of the ways we define a state is by its monopoly on violence. Given that speech is now a form of (real, not metaphorical) violence, it follows that the state should have a monopoly on speech. This is a rather chilling idea, but it fits in well with the repeated advice I receive from identity politics adherents when disagreeing with them: that I need to “educate myself”, apparently until I agree with their propositions. I was told that my use of the words “I think” (among others) indicated that I “hadn’t made an effort to understand”. Re-education camps are a favoured tool of communist, anti-liberal regimes. Thinking, not so much.
• Following this, I stumbled across an identity politics claim that said framing “the issue [of racism] in terms of rationality and irrationality; [is] another trope of racism”, with a rather bizarre quote saying that racism may be rational because it is profitable, which, apparently, makes the converse true: rationality can be racist. The same “logic” could be applied to criminality – if expected (personal) benefits exceed costs, committing a crime is “rational”. It doesn’t follow that rationality is therefore criminal. My (lay) understanding of ethics and rationality is that they’re quite different things. The idea that all people have some inherent worth isn’t rational, or irrational, it’s a moral view. And morals are not derived from reason.
• Apart from shared facts, it also denies the idea of shared values and interests as basis for political action. The primary basis is group identity, one can only form “coalitions” with others based on values, one cannot form “groups” with them. (Actually the Combahee River Collective goes much further than this, denying the possibility of certain coalitions. They include this quote “I haven’t the faintest notion what possible revolutionary role white heterosexual men could fulfill, since they are the very embodiment of reactionary-vested-interest-power”. At least one white heterosexual man – a chap called William Wilberforce – might be able to offer some help answering this question)
• Following from this, an individual must primarily be seen first as a certain race (or skin colour). This is my personal definition of racism – being blinded to the individual by some “group” attribute. It’s tribal, and it’s highly illiberal because individual identity is subverted to some group identity.
• Attempts to share the experience of other groups (e.g. doing yoga, eating thai food, writing about characters from another group) are branded as “cultural appropriation” (imperialist, colonialist, etc). This makes a taboo of the most basic way humans empathise and identify with one-another (by sharing experience, by imagining oneself as “the other”).
• Attempts to talk about issues concerning other groups becomes “speaking on behalf of”, “making voiceless”, “speaking over”, which imposes a form of censorship – certain areas of speech are reserved for certain groups.

Even if one had no problem with any of the above, it remains for identity politicians to explain “how it would work” in practice.

One of the claims of identity politics is that our parliaments should look like our streets; i.e. reflect the demography of the country. Many liberals might agree with this idea, not because it is a desirable end in itself, but because a liberal society has no (or few) structural barriers to opportunity based on race /ethnicity, and so absent barriers, our parliaments might reflect our demography (assuming no generalised differences in preference or aptitude for politics between minority groups). In the US, minority representation in congress and the senate lags behind the demographics, but it is steadily growing (I think the lag is about 20 years, so congress today reflects the ethnic make up about 20 years ago).

Identity politics doesn’t want equal opportunity – or at least, not just equal opportunity. It requires equality of outcome, because one cannot be represented by anyone other than a member of your racial group. On the other hand, one can be represented by someone far wealthier, from a completely different cultural background (e.g. a Somali Muslim background representing someone from a Basotho Christian background), or a different level of education, with different moral values. As long as your skin colour is the same then the most important values are aligned. I’m not sure how this approach is supposed to work in a representative democratic system. If I am black, I cannot be adequately represented by a white local MP – that is an axiom of identity politics. Do all black people therefore have to move into the electorates with black MPs? Is that really the outcome that is going to remove all structural inequality? Or does any electorate with a black voter require a black MP to avoid disenfranchising that voter? Can representative democracy ever pass muster for identity politics?

—

Kimberlé Crenshaw’s work makes for compelling reading. She clearly explains how the intersection of two oppressed groups (blacks, and women) makes for a specific set of needs which are poorly represented by the separate anti-racist and feminists movements. Indeed because black women are marginalised within both of these wider political movements, they can find the wider movements acting against their specific interests. i.e. anti-racism movements better represent black men, while feminist movements better represent white women. It’s a good read.

The intersectionality logic is essentially that sub-groups can be poorly represented by wider groups, which strikes me as an obvious truth (not wishing to detract from Crenshaw’s insight of course!). But what is confusing is that identity politics advocates, who seem very interested in the notions of intersectionality, don’t actually take the logic any further than “black women” (say). I mean, does the group “black women” represent a monolithic set of interests? I mentioned the Combahee River Collective opinion on the viability of a lesbian subgroup (rejected). Is it not conceivable that within your selected “group identity”, intersectionality continues to apply along other delineated differences, such as religious affiliation, sexuality, age, marital status, disability, socio-economics, etc, even one’s position in a single family power structure, until the logic of intersectionality eventually takes us back to the original liberal position, which is to consider the individual, not the group ?

—

My view – as you may have guessed – is that identity politics is illiberal. It limits the scope of possibilities for the individual by defining who individuals must be according to some group status. Just as banning homosexuality is illiberal, because it denies people the ability to be who they really are, a world view which defines people as primarily members of a gender or race reduces their identity to a single dimension, and thus their scope of personal freedom to explore other aspects of identity, behavior and expectation.

Supporters of identity politics might argue that these restrictions only apply to the “dominant group” (the inevitable culprit being the “cis white male”). Obviously they are not suggesting that black people eating pasta, or wearing “european style” clothing is “cultural appropriation”. But I don’t think you can easily establish this one-way behaviour control on the dominant group. As soon as you tell the dominant group that they have “appropriated” yoga in some neo-colonial venture, you automatically put “doing yoga” back into the realm of expected behaviour for people of south asian descent. That’s what delineating difference does.

If such actions “create safe spaces”, they also create safe prisons.

The idea of “cultural appropriation” is problematic for other reasons. Cultures have always borrowed (or “stolen”, if you prefer) from one another, and will continue to do so. Europeans “appropriated” numbers from Arabs and Persians, who had previously “appropriated” them from Hindus. Arabs “appropriated” large chunks of Islam from Jews and Christians, and during the centuries that Arabs were colonising Europeans, or being colonised by them, a huge cultural exchange has taken place. And it’s not like it took place on equal terms, it was usually in terms of oppressor and oppressed.

Does the process of “appropriating” the culture of the “other” work towards their oppression, or towards their social inclusion (and liberation)? Identity politics would have this as a one-way street in favour of the oppressors. I think the reality is much more complex.

More evidence of the two facets of “delineated difference” is offered by a recent twitter-storm over a US figure skater’s medal win. The tweet “Immigrants: they get the job done” was quickly condemned as “perpetual othering”, because the figure skater in question was born in California (to immigrant parents). Now it seems to me that you cannot, on one hand, develop political action based on a (minority) group identity, but at the same time be outraged at “perpetual othering”. Perpetual othering is, in fact, your political project.

Similarly, Reni Eddo-Lodge’s viral post deplores the emotional disconnect of white people who have “never known what it means to embrace a person of colour as a true equal, with thoughts and feelings that are as valid as their own.” Yet identity politics makes the specific claim that the validity of (certain) thoughts and feelings depends on the colour of your skin. There can be no equal validity of thoughts unless the thoughts are expressed from members of the same group. Eddo-Lodge makes this point particularly for conversations about race: the only thing worse than white people who deny there is structural inequality are those who accept it but then think one can converse as equals. As for feelings, empathy is forbidden because that supposes me attempting to place myself in your shoes, which is a capital crime in identity politics.

—

It struck me that perhaps this whole debate is one about means and ends. If liberals agree on the problem of structural inequality, then the argument about identity politics is simply one about the best means of solving the problem.

Crenshaw makes this point too: by saying that “delineating difference need not be the power of domination; instead be the source of social empowerment and reconstruction”.

I think seeing people primarily through the prism of their race is an abhorrent way to live and treat others. Essentially it means adopting the same world-view of white nationalists and supremacists, in the name of freeing minorities from oppression by these same groups. A sort of “fighting fire with fire” approach.

Except that it is not fighting fight with fire, it is just lighting more fires. It validates the world-view of the white supremacists and thus strengthens them. I think it is striking that when the River Collective statement was drafted, and also in Crenshaw’s article, the idea of a “white group identity” is dismissed as impossible or nonsensical (Crenshaw says “… whites do not constitute a specific cultural group” while the River Collective noted that “Our situation as Black people necessitates that we have solidarity around the fact of race, which white women of course do not need to have with white men, unless it is their negative solidarity as racial oppressors”).

But now, we have Steve Bannon saying “the same way you are a proud African-American woman, I am a proud white man. What’s the difference between your pride and my pride?”. The modern white identity movement justifies itself using minority identity logic. There is a growing movement of “white nationalism” and left-wing political commentators refer to Trump’s election victory as a victory for “white identity politics”. This confused article in the Guardian makes the point explicitly to criticise white Trump voters, but nevertheless defends identity politics by saying “political elections have always played identity politics”. So, trying to get my head around this, what I understood is that the problem is not the identity politics per se, it’s just the outcome that matters. It’s not the means, but the ends, i.e. whoever wins by using it.  Identity politics used by whites is racism, but used by minorities, it’s empowering.

In 2014 I reviewed a long-winded manuscript which made the identity politics case, but for white people. It argued that white people should look for kinship with other whites, and stop “hating” their own race. That whites should develop a sense of white identity and pride (the book was called “A quest for identity”, before I had ever heard of identity politics). That there was nothing wrong with other races, but we should naturally prefer the company of our own, just as the other races do. Mixed-race marriage was an anathema to the writer of this book – a betrayal of one’s race. On that subject, I wonder where mixed race marriage fits in the views of the Combahee River Collective, or identity politics in general ? Surely it throws a spanner in the works. The Collective statement demands “solidarity around the fact of race“ and sees no role for white heterosexual men, so I can’t imagine mixed race marriage would go down very well. If this is the case, identity politics is just a rather convoluted theory for ethnocentrism. And come to think of it, referring to the “fact” of race is an interesting choice of words. Usually anti-racists argue that race is a social construct, with no (significant) genetic basis.

This white ethnocentrism is the flip side of minority identity politics. It takes the “minority group” reasoning and applies it to whites. The logic is exactly the same: “each to their own”. And in the book I reviewed, “white identity” was just a thin positive veneer applied to the author’s very negative view of non-white people.

To me it seems fairly obvious that the “means” of minority identity must necessarily have the unintended, but logical corollary of an “ends” which nourishes white identity. Insisting that we see people primarily through the prism of race, that values and political projects cannot be shared across racial groups, was always going to do this.

However even if we think that identity politics will somehow achieve its desired aim – the end of structural oppression, we should keep in mind that ends are uncertain but means are not. Gandhi put it better:

There is no wall of separation between means and end. Indeed the Creator has given us control (and that too very limited) over means, none over the end.

I saw a talk by Eddo-Lodge recently, and she admitted that a society where we were all perceived simply as members of the human race (i.e. the concept of race-as-skin-colour didn’t exist), was a desired outcome, but we were nowhere near that at the moment. I’m not sure how identity politics, which insists on defining people primarily by racial group, brings us closer to this goal. It brought to mind another quip, often (falsely*) attributed to Gandhi: “be the change you want to see in the world”

Identity politics is not the change it wants to see in the world.

—

* he didn’t actually say this, but something more complicated: “If we could change ourselves, the tendencies in the world would also change. As a man changes his own nature, so does the attitude of the world change towards him. … We need not wait to see what others do.”