Thirty-eight years ago, I published a book on John Rawls' A Theory of Justice -- the first of what would be a flood of books about what is arguably the most widely read and influential piece of political philosophy to appear in English in the last century and a half. Next month, I shall start teaching an informal reading group on that work to some graduate students in the UNC Chapel Hill doctoral program, and in preparation for that gig, I have been reflecting on Rawls' argument. I have been elaborating in my mind, as I take my morning walks, a line of analysis that is not present in my book, and which, so far as I know, has not found its way into the literature on Rawls. Since I wanted to spring it on the students, and since I long ago learned that anything posted on the Internet is sure to be read by all the wrong people, I have been keeping it to myself, but having run out of topics to comment on this summer, I have decided to lay it before you in a lengthy post. If the graduate students find their way to it before we formally convene, so much the worse for me.
A word of warning: Rawls first announced his theory in a journal article published in Philosophical Review in 1958. He subsequently made several major revisions in the theory, initially in another article, Distributive Justice, and then in the book, which was published in 1971. Thereafter, he spent several decades tinkering with the theory. None of these changes in any way affect the idea I am going to set forth here, and I shall therefore simply ignore them.
The centerpiece of the theory is the two principles for the general regulation of society that, according to Rawls, would be unanimously chosen by individuals engaged in what Game Theorists call a Bargaining Game. Here is the passage in which Rawls first introduces those principles:
The conception of justice which I want to develop may be stated in the form of two principles as follows: first, each person participating in a practice, or affected by it, has an equal right to the most extensive liberty compatible with a like liberty for all; and second, inequalities are arbitrary unless it is reasonable to expect that they will work out for everyone's advantage, and provided the positions and offices to which they attach, or from which they may be gained, are open to all.
The central idea of these principles is this: Modern society consists of large-scale bureaucratic organizations -- corporations, universities, hospitals, government offices, armies, and so forth -- in which there are clearly defined roles to which attach specified duties and compensations. The default baseline situation, we may imagine, is one in which all the roles receive the same compensation -- a situation of absolute equality. However, if the ablest individuals with the best sets of talents and skills are drawn into certain key positions, the institutions will function much more efficiently -- to be put it as simply as possible, the net output of the institution will be higher. Now, to ensure that these especially talented or well-prepared individuals end up in those key positions, it is necessary to pay them higher salaries, and this is an element of inequality. But the gain from their greater efficiency will be such that after they receive their added compensation, there will be something left over that can be given to everyone else [or to the least advantaged representative individual, depending on which version of Rawls' theory we are considering.]
Let us call this something extra the "inequality surplus" [not Rawls' term, by the way.] Assuming, as Rawls does, that the individuals in the society are "not envious," which is to say assuming that those getting less in the way of compensation than the key individuals do not begrudge them their higher salaries so long as they themselves are getting more than they would without the productive efforts of those key individuals, everyone will endorse this system. And that, in a nutshell, without all the baroque elaborations, is Rawls' argument.
Since this is rather abstract, let me restate it by way of a hypothetical example. Consider a manufacturing firm that makes washer-dryers. The employees, we may suppose for simplicity's sake, are divided into executives who direct the operations of the firm, production line workers who assemble the washer-dryers from components delivered to the factory, and loading dock workers who unload the components when they are delivered by truck to the back door of the factory and load the finished washer-dryers onto trucks waiting to take them to retail outlets.
Clearly, the earnings of the company will be much greater if those with special skills, training, and talent for corporate management are assigned to the executive jobs, and that fact will make it possible to raise everyone's salary. But there is a problem. Rawls does not identify this problem, but his theory makes no sense at all unless we assume that the problem exists, so we are justified, I think, in assuming it.
The problem is this: After the professionally administered aptitude tests are scored, and the individuals with special management talents are identified, they are offered jobs as managers. But when the selected individuals are invited into the executive suites, they say, "Thanks, but no thanks. I would rather work on the loading dock."
"What is this?" you say incredulously. "Where on earth does Rawls say that in A Theory of Justice?" Well, nowhere of course. But he must be assuming it, even though he doesn't know it, because if those tapped for management actually prefer to be in management [or, technically, are indifferent between executive suite and loading dock, but never mind that], WHY PAY THEM MORE TO TAKE THE JOBS?
"But it is not just to pay them no more than loading dock workers, and Rawls says his theory is a theory of justice!” you protest. "Ah," I reply, "you have not read Rawls as carefully as you ought. Rawls does not start with a pre-systematic concept of justice that he assumes without argument. He starts with a collection of rationally self-interested individuals who, according to him, will out of self-interest choose these two principles, and the fact that they will out of self-interest choose these principles MAKES THEM the principles of justice."
There is no reason for me, a rationally self-interested individual, to approve a system of unequal compensation unless I believe that doing so will draw into key positions individuals whose greater efficiency will end up benefitting ME.
Does anyone at all really believe that offered a choice between corner offices in the executive suite and nine-to-five jobs on the loading dock, potential executives will opt for the loading dock unless they are paid hefty salaries well above that of their lesser brothers and sisters out back? Rawls does. He must. Otherwise the centerpiece of his theory collapses.
Now, it is entirely possible that extra compensation will be required to draw people into some important jobs. Maybe those with a gift for medicine will choose that career from sheer love, but would anyone actually want to be a corporate lawyer if it weren't for the pay?
Think about my own profession, university teaching. Since universities are [or used to be] non-profits, it is a little hard to translate the gain from selecting talented philosophers for the professoriate into dollars and cents, but clearly there is some social benefit to having Plato or Kant taught by folks who actually understand them. Now, a university Philosophy department employs a bunch of professors and some secretaries. I am vain enough to believe that the education students get is actually better for having me teach the classes rather than having me sit in the front office and answer the phone. But if I were offered both jobs, would I choose the secretarial position unless the professor's slot came with a salary bump? Of course not. I enjoy teaching much more than I would enjoy answering the phone, even taking into account grading papers.
There is much more to be said, but this post is long enough, so I shall conclude by simply noting that A Theory of Justice, when all is said and done, is an ideological rationalization of mid-twentieth century American welfare state liberalism. Think about it.