Recently, a friend of mine (incorrectly) told a new acquaintance of mine that I didn’t “believe in privilege.” As my new acquaintance began explaining how she couldn’t imagine how anyone could deny the existence of privilege, I decided to think more precisely on what privilege is, whether it exists, and who has it. I’ve now decided to write this post to formalize the idea of privilege and see where a consistent application of the logic behind it leads.
A good starting point for defining privilege is to describe it as a condition in your life that you did not pick but that had a benefit for you. For example, white people tend to, on average, have higher salaries than black people. White people usually did not become white of their own efforts, yet they reap the benefits of this physical feature. Why? Some reasons are that they have historically been more free to accumulate capital, have not been the subject of race-based oppression by the government and system, and have been a majority in the US for a long time. These things often lead to higher confidence, higher income, higher levels of education, more positions of political power, etc.
One way of formalizing the idea of privilege, then, is to ask “All else being equal, does having this characteristic mean that I will be more likely to be successful?” Example: Take a person born into a middle-income family in a middle-income neighborhood. If we let the person live his entire life, then went back in a time machine and changed only his race, would this have led to a better/worse life outcome?
The next step to take is to notice that sometimes our characteristics can take on a variety of values, such as race, instead of being a binary “yes/no.” For example, a person can be white, black, Hispanic, Asian, or any number of sub-categories of these and other races, as well as a mix of various categories of races.
Therefore, we begin to see privilege as a relative concept that can be ascertained given two possible values for a specific characteristic: “All else being equal, for a given personal characteristic ci, does value cib create a more favorable outcome than value cia?”
At this point we need to add another refinement. It seems like a good idea to define privilege in relation to a specific goal that an individual has. If a person doesn’t care about total life income, having a value for a personal characteristic that only increases life income relative to alternative values for that characteristic doesn’t matter – the person doesn’t care about life income anyway. Similarly, if a person has a specific underprivilege that would only affect a goal about which he doesn’t care, then this is not a disadvantage to him, and he shouldn’t be considered to be underprivileged .
Given the goal-directed nature of privilege, the next step is to realize that since individuals have a variety of goals, some often polar opposites of other people’s goals, a certain value for a characteristic could be
– a privilege for one individual
– an irrelevant characteristic for another
– an underprivilege for a third
Insofar as a set of goals of individuals tends in the same direction (loosely speaking), a given value for a characteristic that is useful will tend to be a privilege for each individual, and hence can be termed a mark of privilege for the group. That is, it makes sense for two individuals in that set to look at each other and judge how privileged each is compared to the others by the values of their numerous characteristics and how they affect the achievement of the goals they all happen to share.
Let’s list some examples of privilege in the US. Each item assumes that all members of the society (the US) have identical preferences. I give some caveats in parenthesis to show examples of when an item would not be a privilege, either because there is a reasonable chance of a preference for a goal not being identical among individuals or because the specific value of the characteristic in question would not be conductive to achieving the shared goal:
– Being white (caveat: your words as a white person might hold less water when trying to convince a group of black people of something in some situations, all else being equal)
– Being male (caveat: if one of your main goals in life is to spend time with your children and you’re going through a divorce, it’s more likely your ex-wife will retain custody, your masculinity here being an underprivilege)
– Being intelligent (caveat: in some circles, intelligence might be viewed with suspicion as a sign of eliteness and superiority)
– Being physically attractive (caveat: beauty is in the eye of the beholder; being attractive to one group in society doesn’t guarantee being attractive to another)
– Higher family income (caveat: you might lead a miserable life [by your own judgment] if your parents only care about money)
This list shouldn’t be too controversial. In most cases, having one of the above qualities will, all else being equal, make life nicer for you, and should be considered a privilege.
Now, let’s return to the definition of privilege. It’s useful to distinguish between goals you’ve achieved thanks to your privilege and goals you’ve achieved thanks to you merit. Therefore, success can roughly be modeled as a function S=f(P,M) – that is, a function of privilege and merits. The privileges are the environmental variables that made you more successful, all else being equal. Your merits are the efforts that you put in independent of your privilege. Keep these two categories in mind as we dig deeper to see whether this distinction is at all sustainable.
Let’s look again at the S function: S=f(P,M). P is some combined measure of various types of privileges a person possesses, perhaps a vector in Rn. Once again, each privilege value is defined in respect to a goal an individual wants to achieve . M is some measure of how much effort a person is willing to put in to achieving success, and hence the level of merit of the person. I list M here as scalar, although it might very well be a modeled as a vector in Rm. It’s reasonable to say that this function is monotonically increasing in each variable, or that the partial derivative of S is positive with respect to either variable. That is, keeping merit constant, the more privilege you have, the more successful you will be (by the definition of privilege). Also, keeping privilege constant, the more meritorious you are, the more successful you will be as well.
We can now determine your marginal “success-productivity” of privilege (MSPP) and the marginal “success-productivity” of merit (MSPM). The MSPP tells you by how much your success will increase if you keep merit constant and increase privilege by one unit (don’t worry about what these “units” are – we’re discussing general intuition here). The MSPM tells you how much your success will increase if privilege is kept constant and merit increases by one unit.
The key idea to take away from this seeming over-formalization is that the marginal effectiveness of being more meritorious can depend on your current total privilege. Similarly, the marginal effectiveness of being more privileged can depend on your current total merit. I will give an analogy from economics to explain what I mean:
In economics, the output of a firm is a function of its quantity of capital and its quantity of labor. For example, Q=bKaL1-a, where Q is output, a and b are some constants, K is the amount of capital, and L is the amount of labor. (Note: this specific form of the production function is called a Cobb-Douglas production function). Now, intuitively, say that a person is willing to invest 10 days into making cars. If the capital he has at his disposal is no more advanced than that available in the mid-19th century, he might be able to produce one car in ten days. If, on the other hand, he has very modern capital including various kinds of powerful and versatile machines, in 10 days he might be able to produce 3 cars by himself. We see from this example that the marginal productivity of his labor (in this case the additional cars he can make with each additional 10 days of work he puts in) depends on the capital he has available. The more capital (or, in this case, the higher quality the capital), the higher the productivity. This can be seen if the partial derivatives of Q are taken with respect to K and L – the higher the K, the higher the benefit of adding one more unit of labor. The higher the L, the higher the benefit of adding one more unit of capital.
Moving back to privilege, we can see how this applies: If your environment is very poor, increasing your personal efforts (separate from your success due to privilege) will result in small increases in success. If, on the other hand, you are very privileged, each additional unit of merit has a high impact on your success. Similarly, if you are not very meritorious, adding more privilege won’t help as much as if you’re very meritorious, all else being equal. 
Now, let’s try to further refine the definition of privilege. We stated before that factors like family income confer privilege. These are environmental factors external to the choices a person makes that affect his success. Let’s look at the privilege of being broadly considered attractive. This is a privilege with which you were born, and it exists thanks to the good genes of your parents and also thanks to pure chance. Sure, perhaps you spend a lot of time working out, dressing well, or whatever else your society values, but there is some degree to which your attractiveness is a privilege by birth – that shouldn’t be controversial.
To better explore the limits of privilege, let’s look at the other alleged component of success: merit. How can we nail down merit? Well, merit, which may loosely defined as the effort a person puts in, is (by construction) a variable separate from privilege. But can merit ever actually be separated from privilege? After all, how industrious and hardworking a person is is due in part to their education. And since education is very much affected by family income, merit becomes entangled in privilege. Furthermore, the effort one puts in also depends on the values instilled in him by his parents. This, too, was out of his control, and is a part of privilege. Parents who care about their children will pass on better life lessons. The third big factor affecting one’s effort in life is his genetics – both his propensity toward hard work and the IQ that is conferred by his genes. The person did not choose his genes. In fact, from our discussion of the privilege of being attractive, we saw that genetic traits are a form of privilege.
We see that all the factors that affect merit (“how much effort, work, and sweat a person ‘chooses’ to invest”) are in fact a type of privilege. If merit is a function only of privileges, then there is no real category of “merit” separate from privilege. All the efforts you’ve put into life have been a product of either nature or nurture. In either case, they’re privileges (or underprivileges) that were conferred to you externally.
We arrive to the conclusion that pretty much everything is a privilege and no one deserves anything due to “merit.”
Lest you think this conclusion too radical, consider the following thought experiment:
Imagine taking a person who, in 2014, was upstanding, smart, tolerant, respectful, and considerate, and magically making him be born in a war-torn nation in the ancient world populated by mostly racist and ignorant people. This person would likely not have turned out tolerant, respectful, and smart. If in 2014 he could have learned integral calculus and the intricacies of DNA replication, in the ancient world he would have believed that storms are caused by the gods and that the sun goes around a flat Earth. This thought experiment makes it apparent just how strongly who we turn out to be is influenced by our environment (which, insofar as it is conductive to our goals, can be called our privilege).
Our conclusion so far has been that every single part of your life that led you toward success has been a privilege, and that you didn’t really make any choices that made you “deserve” your success in any moral sense. This discussion about privilege, then, begins to look very much like a discussion about determinism, and I’d argue that privilege could be just a face of determinism – the one that specifically deals with people’s success. Since we saw that success in life is fully determined by privilege, it is natural to make the symmetric argument that failure in life is also determined by privilege – or, in this case, the lack of privilege, or underprivilege. That is, if something very bad happens to a person, it is because they were underprivileged.
Now, let’s analyze privilege under two different scenarios of assault:
1) A perfectly rational man living in a world with no uncertainty one day cuts off the foot of another man. He is sentenced to death for this. Question: since his action led to a state of being that any normal person would deem unfavorable (ie, being executed), and since we concluded that failure is just as much the result of privilege (and is due to no merit or dismerit of one’s own), does it make sense to say the attacker is underprivileged? After all, his environment conspired in such a way as to make him end up being convicted. And just like successes are not due to one’s merit, the same goes for failures. The man who was attacked, on the other hand, continues living. His life is doubtless much worse for having lost a foot, but at least he is alive. I doubt it’s easy to argue that being attacked but alive is worse than being dead (though I suppose, in my ignorance, that it could be possible). Hence, it makes sense to actually ask whether the man who was attacked was more privileged than his attacker, who will go on to be killed. However, in this case, the answer must necessarily be “the fact that he will die is not enough to show he is less privileged, despite death being really bad.” Since the man is perfectly rational (by construction) and there is no uncertainty in the world, his action maximized both his ex-ante and ex-post utility (which I will use interchangeably with success). Therefore, before he committed the act, he knew perfectly, 100% well what would happen and chose to do it anyway. This tells us something about his preferences: the act was to him more valuable than life itself. Given that he chose an action that he valued more highly than all the possible satisfaction of life, it appears that the criminal may be much more privileged than the victim. Of course, there exists the possibility that this isn’t the case, but it is not easy to show that this would have a high probability a priori.
2) A man with bounded rationality and imperfect information one day cuts off the foot of another man. He is sentenced to death for this. In this case, does the same analysis as before apply? Notice we have removed the assumption of perfect rationality. Now, it could very well be that the man was very irrational and couldn’t properly assess the consequences of his actions. Moreover, he had imperfect information, and maybe he had no idea that facing death in an hour would be so awfully terrifying. The attacker goes on to lose his life, which may reasonably be assumed to be very tragic for a person. Once again, his actions were due to his environment/privilege. In this case, since his environment conspired to lead to an outcome where one of the worst things that could happen to a person did indeed happen, his environment conferred upon him severe underprivilege. The man who was attacked is, once again, alive but very distressed by the occurrence. Yet he still can go on to have a meaningful life. (Before you disagree with me for being callous, consider what disagreement with that statement would mean: that people who get disabilities can never have meaningful lives after the incident). Hence, the man is in a better position to achieve success, happiness, and satisfaction than his soon-to-be-dead attacker is. As such, by definition he is more privileged than his attacker. Why? Once again, because the actions of each individual are wholly determined by environment, and privilege exists when the environment is such that the events in one’s life result in success, relative to the alternative. Since the victim is able to have more future satisfaction than his attacker (or present discounted value thereof), he is more privileged than the attacker. Of course, it could be that the victim could go on to have a miserable life, in which case it can be argued that no, his attacker was in fact more privileged. This is true – but I am talking about the range of possibilities.
We now come to the topic of telling people to check their privilege. When one checks his privilege, this can be described as him exploring the privileges that led to his present (and future) success and seeing their relative contributions toward this success. He also compares his levels of privilege to either those of the person asking him to check his privilege or to those of a group which the person asking him to check his privilege is discussing. I can see a few legitimate reasons for someone being told to check his privilege, which I shall label as Types:
Type 1) “Check your privilege: You’re failing to realize how my/the group’s lack of privilege in some dimensions is leading to my/their lower success. By doing this, you are making me/them feel bad, which is bad for me/them.”
Type 2) “Check your privilege: You’re failing to realize how my/the group’s lack of privilege in some dimensions is leading to my/their lower success. By doing this, you’re failing to understand causal pathways in society, leading to an underdevelopment of your own understanding of how society functions, which is bad for you.”
Type 3) “Check your privilege: You’re failing to realize how your privilege in some dimensions is leading to your higher success. As such, you overestimate the privilege of others, you think that others are undeserving, and hence you do not support efforts (voluntary/coercive) to improve their success.”
Is it useful for you yourself to check your own privilege? It can indeed be in cases 1 and 2 – when failing to do so either makes other people feel bad or when it prevents you from understanding your own source of success.
Take the (contrived) example of a restaurant owner in a middle-class neighborhood that has some percentage of its patrons be poor families. Say this restaurant owner dislikes poor people and takes any chance he has to poke fun at them for not being successful. In this case, he might very well alienate patrons and have them stop coming. If he checked his privilege, he could become more understanding of their troubles and stop poking fun at them. They wouldn’t leave, and he would have more patrons.
Another example is that of the entrepreneur. If he understands that a certain social structure creates a cycle of poverty, he could try to start a firm that sells a product that helps people escape poverty (by, say, increasing people’s human capital). By checking his privilege and understanding the causes leading to poverty, he is able to create a product to circumvent these social circumstances.
I will have more to say on the topic of why privilege checking can be of utmost importance – even to people who would consider themselves “conservative.”
In summary, we have learned a few things:
– Privilege exists relative to the goals of an individual.
– Since the importance of goals to various people cannot be compared (interpersonal utility comparisons are impossible), subjective privilege cannot in fact be compared.
– If we ignore this important theoretical point, we may say that insofar as people’s goals are the identical, everyone has privilege relative to an alternative value for a given characteristic.
– The social justice discussion on privilege is, at its core, an argument about the deterministic nature of the universe. Privilege is one face of determinism.
– Just as privilege reduces the emphasis on personal merit in achieving success, it similarly reduces the emphasis on personal fault for failures – including crimes against victims.
– As Peggy McIntosh points out in her seminal paper on privilege, privilege is not necessarily bad. In fact, I would argue that the goal should be to increase everyone’s privilege as much as possible.
– Analyzing events in the framework of privilege can give us odd conclusions – such as the fact that in a framework of bounded rationality and imperfect information, a man who is made disabled by another may in fact be more privileged than his attacker. Such language is likely foreign to the discussion of privilege that exists among those interested in social justice, but is one that follows from our careful definitions.
Impact of conclusions
It is not my purpose to tie the arguments presented here into a larger philosophy in order to tell people how they should change their actions. I merely sought to follow the logic of privilege to its conclusion. Whether the framework itself is accepted is another matter. Still, I will note two impacts of my analysis:
– The existence of privilege cannot be denied by any reasonable person. The fact that you are far more successful in 2014 America than 2014 Congo, 1900 America, or 1900 Congo is stark enough evidence. Your place of birth significantly affects your prosperity.
– The importance of privilege analysis will rise and fall with the importance society places on an acceptance of a deterministic view of the world. If society decides that determinism shouldn’t affect how we view the moral positions of various actions and individuals, it logically should also not assign much importance to privilege analysis. On the other hand, if much weight is placed on determinism, much weight should be placed on privilege analysis as well. Conversely, if society decides privilege analysis is important, it should consider all the implications of determinism – for they are part and parcel of privilege analysis.
Much of this article makes arguments of the form “if X is accepted as valid, Y follows.” This is not meant to be interpreted as “X is valid, and therefore Y is true.” Furthermore, I have not passed moral judgment on almost any topic related to privilege. For example, saying that privilege analysis in the case of bounded rationality points to the possibility of a man who is attacked being, overall, more privileged than his attacker does *not* imply that I think that his attacker is a more moral person than the victim is because he is less privileged. I simply state he might be less privileged, whatever the moral implications of that are. This stems from the definition of privilege. In fact, the hypothetical scenario I discuss and its conclusions on privilege are one of the reasons I find privilege analysis to not be a very good framework of normative analysis (normative specifically – it is still useful for positive analysis).
I’d like to thank my co-blogger for taking the time to slog through this post and offering constructive advice. Any hatemail you have should be directed to me, for the ideas presented stand behind my name.
 Once we see privilege as being evaluated within the means-ends framework, we need to take into account the fact from economics that interpersonal utility comparisons are not possible. That is, even if a person has a privilege that objectively allows him to achieve a certain goal more easily than a person without that privilege, and even if the two people have an identical rank of their goals, it cannot strictly be said that the first is more privileged, since utility cannot be compared among people. Therefore, it is necessarily correct to say that privilege is actually *unable* to be compared across people when privilege is couched in terms of subjective valuations of ends. In the remainder of this article, I will ignore this and pretend that this utility among individuals can in fact be compared, despite believing that this is not possible even in theory and that it doesn’t make very much sense. I do this because most people do this subconsciously, and I will play along, ignoring the majority of the economics profession.
 The implications of the functional form and of the form of the variables P and M allow the model to generalize to intersectionality of privileges and interlocking hierarchies. For an intuition, see the economic analogy to capital and labor in an economy.
 I will add here that it is possible for the function f() to have a negative second derivative – that is, for your success to increase with both P and M, but more and more slowly over time. That is, the marginal benefits of increases in P and M go down as you have more of both P and M. An example would be increasing your merit in a very capital-poor society. No matter how meritorious, you likely will never be able to send a spaceship to the moon.