John Rawls: “A Theory of Justice” versus his later works

This essay was part of the second year of my MA in Philosophy and concentrated on the work of one philosopher exclusively: John Rawls. If you haven’t read any of his work, I would advise against reading this essay. There will be another essay about the philosophy of punishment which should be of greater interest.

This essay is supposed to compare two publications by John Rawls, his seminal book A Theory of Justice1 and his essay The Domain of the Political and Overlapping Consensus2 and to determine if the position outlined in the latter marks a fundamental change from the views expressed in the former.


The question should be easy to answer as both texts were written by the same author and The Domain of the Political and Overlapping Consensus includes a chapter VIII3 explicitly entitled “Comparison with A Theory of Justice”. However, it will become clear from going through this short4chapter that its title feigns support and helpfulness which are woefully absent and that Rawls’ remarks about his own works yield less clarification than hoped for.

Not an easy read.

Rawls begins by stating that “it may seem that the idea of an overlapping consensus … [is] a significant departure from [A Theory of Justice]”5 and that there is “some departure certainly”6. Rawls states here that A Theory of Justice “emphasizes the limited scope of justice as fairness”7 and that the book “never discusses whether justice as fairness is meant as a comprehensive moral doctrine or as a political conception of justice”8 and indeed that it contains “no mention of the distinction between a political conception of justice and a comprehensive doctrine”9 at all.

Where Rawls becomes more concrete in the chapter “Comparison with A Theory of Justice”, he calls parts of A Theory of Justice “utopian”10 and not realistic11. In Rawls’ later view, “the idea of an overlapping consensus was … introduced to think of the well-ordered society of justice as fairness in a different and more realistic way”12. This “overlapping consensus exists in a society when the political conception of justice … is endorsed by each of the main religious, philosophical and moral doctrines”13. Rawls expressly states that “we can no longer assume that citizens generally, even if they accept justice as fairness, also accept the particular comprehensive view in which it might seem to be embedded in [A Theory of Justice]”14. In contrast, and in one of the few examples of clarity about the difference between the two works, “we assume their overall view has two parts”1516, the political conception of justice for once and – possibly related to the first in some manner17 – a comprehensive doctrine as the second part18. When it comes to the relation between these two parts however, Rawls sadly retreats into vagueness again: “It is left to citizens individually to decide for themselves in what way their shared political conception is related to their wider and more comprehensive views.”19

Rawls’ concluding remarks in the chapter “Comparison with A Theory of Justice” make it seem as if his overall idea of what a fair or just society should like have not changed, but that only his argumentation has: “This is a better and no longer utopian way of thinking of the well-ordered society of justice as fairness. It corrects the view in [A Theory of Justice], which fails to take into account the condition of pluralism to which its own principles lead.”20

Although it is not up to me to deny Rawls the criticism of his own earlier work, I find it too harsh to call it “utopian”. A Theory of Justice made it clear that the “veil of ignorance” and the “original position” were (only) thought experiments21 and not to be interpreted as a description of a present or desired society. Based on my reading of A Theory of Justice, I found it convincing. I did not see it as an inadequate theory that called for correction or improvement.


Matravers lists “two main differences between early and late Rawls”22: First that The Domain of the Political and Overlapping Consensus “focuses on ‘the domain of the political’ and takes the two principles of justice as ‘an23 example of a political conception’”24. Matravers interprets this to mean that “what Rawls calls ‘the domain of the political’ is broader than justice as fairness and includes it”25. And second that “Rawls is explicit that his account is grounded in conditions that will only be found in a ‘well-ordered democratic regime’”26. Matravers suggests that the overlapping consensus “would root Rawls’s account firmly in political fact … with the result that it would not be a theory of justice for all people at all times, but a theory of justice only for a particular sort of pluralist, democratic society.”2728

“I bet I can confuse you.”

This later point is something which I found quite disturbing in The Domain of the Political and Overlapping Consensus. Rawls states that “those who grow up in a society well-ordered by [justice] … develop a sufficient allegiance to those institutions”29.

I find this circular. It is a more convoluted way of saying “living in justice produces a sense of justice” or – to exaggerate – “you will recognise justice when you see it”. This is not only methodologically unsatisfactory as it fails to even attempt to explain how justice can be created in a society that is as yet devoid of it, it also takes an enormous step back from the far more universal approach in A Theory of Justice which was applied to all “rational men”30, not only those living in liberal democracies. In fact, in A Theory of Justice, the original position even “specified the political setting that would give rise to the democratic institutions that would be necessary in justice as fairness.”31 In this way, the original position of The Domain of the Political and Overlapping Consensus “has a more modest task”32.


The “key term in late Rawls”33, according to Matravers, is “the concept of the reasonable and reasonable disagreement”34. Wellman also stresses Rawls’ belief that “the solution … lies in ‘public reason’”35.

But in The Domain of the Political and Overlapping Consensus, Rawls is also expressly sceptical about the use of reason. In the chapter “Burdens of Reason”36, he lists a number of reasons why people are not exclusively swayed by reason and also why the use of reason won’t necessarily lead to agreement. After all, there is plenty of room for “reasonable disagreement”37.

This sounded to me far more sceptical and pessimistic as A Theory of Justice in which Rawls assumed that every rational person would accept the original position. Granted, reason is not the same as rationality, but in The Domain of the Political and Overlapping Consensus, Rawls contrasts having reason with being “irrational”38 and as far as I could see, he did not venture to point out the difference in his use of “reasonable” and “rational”.


By “showing that justice as fairness is the most compatible conception”39 and by declaring it to be a “free-standing view”40, “one that is independent of any comprehensive moral theory or doctrine”41, Rawls seems to seek the support of people who would not accept his justice as fairness as their comprehensive view and who might have quite different comprehensive views from his own. I understand him to hope that they can all gather under the big tent of the “overlapping consensus”42.

I doubt however if this method will lead to anyone who doesn’t share Rawls’ egalitarian outlook to change his or her mind. For example, “it is not clear that utilitarians could ever accept Rawls’s priority of the first principle over the second“43.


In conclusion, the statements by Rawls himself44 make it obvious that he thinks there are noticeable differences between his reasoning in A Theory of Justice and the reasoning later displayed in The Domain of the Political and Overlapping Consensus. I see no reason to distrust the author’s judgement on his own works. Also, if there were no differences at all, presumably there wouldn’t have been any need for a new publication.

On the question whether these differences are fundamental I would have to remain agnostic if I had to judge this question on substance alone, not least because I share both Warburton’s description of A Theory of Justice as “a complex and in places rather dry book”45 and Matravers’ assessment that “there is disagreement as to what late Rawls is claiming”46. I would like to add my own disappointment not only about my lack of understanding large parts of what I have read in the process of preparing for this essay, but also at Rawls’ failure to explain the changes in his theories or his thinking and how they affect his conception of justice47. Given the importance of A Theory of Justice in philosophical discourse since its publication, this is an almost wantonly negligent omission. I thought I understood the gist of A Theory of Justice after I had read it; after reading Rawls’ later works, I am not sure about this anymore. “Si tacuisses, philosophus mansisses” comes to mind.

If I had to make an intelligent guess about whether the reasoning of the later Rawls marks a fundamental change from the early Rawls, I would say it doesn’t. For the changes to be fundamental, they would have to imply a discontinuity between The Domain of the Political and Overlapping Consensus and A Theory of Justice. But the former can be seen as a continuation, explanation, expansion, variation of A Theory of Justice. If The Domain of the Political and Overlapping Consensus is a correction of A Theory of Justice, it is only a correction in parts.

I take support for this view from the preface to the revised edition of A Theory of Justice, written by John Rawls in November 199048, and hence after the publication of The Domain of the Political and Overlapping Consensus, in which Rawls writes that “despite many criticisms of the original work, I still accept its main outlines and defend its central doctrines”49. He adds: “if I were writing A Theory of Justice over again, I would not write, as authors sometimes say, a completely different book”50.

This, to me, is sufficient evidence against assuming that The Domain of the Political and Overlapping Consensus was intended to contain fundamental differences from Rawls’ earlier work.

1I use the revised 1999 edition of A Theory of Justice which had first been published in 1971.

2Published in 1989.

3Rawls 1989: 171-2

4Not even one and a half pages long.

5Rawls 1989: 171

6Rawls 1989: 171

7Rawls 1989: 171. The “limited scope“ quote can be found at Rawls 1999: 15.

8Rawls 1989: 171

9Rawls 1989: 171

10Rawls 1989: 172

11Rawls 1989: 171

12Rawls 1989: 171

13Rawls 1989: 176, footnote 1

14Rawls 1989: 171

15Rawls 1989: 172

16See also Vaggalis: 1 and 2

17Rawls 1989: 172

18Rawls 1989: 172

19Rawls 1989: 172

20Rawls 1989: 172

21Rawls 1999: 8; Wellman 2002: 66

22Matravers 2005: 44

23Emphasis added by Matravers 2005: 44.

24Matravers 2005: 44, quoting Rawls 1989: 160

25Matravers 2005: 44

26Matravers 2005: 44, quoting Rawls 1989: 172

27Matravers 2005: 45

28See also Vaggalis: 1

29Rawls 1989: 165. A similar statement can be found at Rawls 1989: 170.

30Rawls 1999: 11

31Vaggalis: 2

32Vaggalis: 2

33Matravers 2005: 44

34Matravers 2005: 44

35Wellman 2002: 69

36Rawls 1989: 162-3

37Rawls 1989: 162-3

38Rawls 1989: 162

39Vaggalis: 2

40Vaggalis: 2

41Vaggalis: 2

42In Rawls 1989: 178, footnote 26, Rawls gives an example in which three different comprehensive doctrines can all agree on justice as fairness as their overlapping consensus.

43Vaggalis: 4

44Rawls 1989: 171-2

45Warburton 2006: 241

46Matravers 2005: 44

47Vaggalis: 3 also calls this “most unsatisfying”.

48Rawls 1999: xvi

49Rawls 1999: xi

50Rawls 1999: xi



Matravers, Derek and Pike, Jon (2003) (editors) Debates in Contemporary Political Philosophy – An Anthology, Abingdon, Routledge

Rawls, John (1999) A Theory of Justice, Revised Edition, Cambridge, Massachusetts, Harvard University Press

Simon, Robert L. (2002) The Blackwell Guide to Social and Political Philosophy, Oxford, Blackwell

Warburton. Nigel (2006) Philosophy: The Classics, 3rd edition, Abingdon, Routledge


Rawls, John (1989) “The Domain of the Political and Overlapping Consensus”, The New York University Law Review 64, 1989, Chicago, University of Chicago Press (reprinted in Matravers/Pike [2003] and quoted by the pages of the reprint)

Wellman, Christopher Heath (2002) “Justice”, in: Simon, Robert L. (2002) The Blackwell Guide to Social and Political Philosophy, Oxford, Blackwell

Study material

Matravers, Derek (2005) Liberalism and Communitarianism, A851 Issues in Contemporary Social and Political Philosophy, Chapter 3, pp. 39-49, Milton Keynes, The Open University

The Internet

Vaggalis, Ted (undated) John Rawls’s Political Liberalism, URL = <> (quoted according to the pages of the A4 printout, comprising 4 pages in total)

About Andreas Moser

Travelling the world and writing about it. I have degrees in law and philosophy, but I'd much rather be a writer, a spy or a hobo.
This entry was posted in Philosophy and tagged , . Bookmark the permalink.

4 Responses to John Rawls: “A Theory of Justice” versus his later works

  1. Pingback: Equality versus Sufficiency | The Happy Hermit

  2. Pingback: Punishment and Liberalism | The Happy Hermit

  3. Pingback: MA Philosophy at the Open University | The Happy Hermit

  4. Um … some interesting points there. Seriously overwhelming, but interesting – what little I really understood!
    I gotta go read about some military hardware now – it helps with the headaches. ;)

Please leave your comments, questions, suggestions:

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s