Erik Hammar

I am lucky to have a girlfriend who understands my academic passions. I recently received a cherished gift from her, a beautiful little edition of Rene Descartes’ brief but worthwhile Discourse on the Method (various title translations; 1637). It is an autobiographical and philosophical work, in which Descartes tracks his record in developing and living according to a certain “method for the well guiding of reason” he has devised. The four maxims which constitutes this method are quite interesting, so let me quote them almost in full:

1. “The first was, never to receive any thing for true, but what I evidently knew to be so”
2. “The second, to divide every one of these difficulties, which I was to examine into as many parcels as could be, and, as was requisite the better to resolve them”
3. “The third, to lead my thoughts in order, beginning by the simple objects, and the easiest to be known; to rise little by little, as by steps, even to the knowledge of te most mised, and even supposing and order among those which naturally do not precede one the other”
4. “And the last, to make everywhere such exact calculations, and such general reviews, that I might be confident to have omitted nothing”

A set of criteria for the method of gaining knowledge, he says, inspired by the step-by-step deductions of the geometricians and mathematicians. These maxims indicate what I think is interesting, that the common portrayal as Descartes as a skeptic and pessimist about knowledge is inaccurate and confused. Descartes thinks, quite simply, that there is quite a lot of things that human beings have the ability to gain firm knowledge about, although he is very aware of the risks of personal and cultural bias and prejudice we always run in doing so. In light of this, I want to elaborate and put into context Descartes’ thought, but I must apologise in advance to those who are experts on Descartes specifically, as opposed to philosophically interested generally – I am no expert on Descartes. My summaries and interpretations will be crude. (I also don’t aim to criticise Descartes’ views, as I am afraid I don’t have time to write that much.) However, what is partly crude need not be fully useless, if it can offer introduction or useful overview.

To remind, Descartes is most famous for his “method of doubt”, expounded most elaborately in his Meditations on First Philosophy, whereby he starts by clearing away all those bits of knowledge that he can conceivably doubt, to see what remains. Sensory impressions, for example, must initially go, since they can be doubted on the basis that we can believe that we have certain sensory experiences when we are dreaming or delirious. What is left, he asks, if I remove from the tables of my mind all that I can conceivably doubt. Famously, he argues that the one thing about the world that he cannot doubt is that he himself exists, since in doubting, he is thinking, and thus, there is something that thinks (cogito, ergo sum).

However, because the summary of Descartes often stops at this point, he has become something (I believe) of the embodiment of the caricature philosopher – the lofty, detached skeptic, questioning how you know you are sitting in a chair and that you are eating müesli, as everyone else gets on with the business of real life. Now, as an aside, even if this were a true characterisation of Descartes’ views, the disregard for it would be up for grabs. Skepticism about your eating müesli might be a defensible epistemological position in some sense, and nothing immediately follows about whether you should disheartedly drop your spoon or defiantly keep crunching your müesli in the face of its ontological dubiousness.

But to return to Descartes, the crucial distinction is that between doubt as philosophical method and doubt as philosophical position. The latter, only, is skepticism. Descartes tries to demolish the apparent foundations of knowledge through the method of doubt, only in the hope to erect a more stable epistemological abode on top of them. Put in its context his attempt is put in a very different light from that of the Philosophy 101 caricature. Descartes is writing in the midst of the intellectual upheaval of the Enlightenment and the ideological debates between established Church and Science. He is looking for a foundation for science, rather than a bizarre skeptical argument for doubting the material world of chairs and tables around us.

His argument, shorltly, is something like what follows. He can perceive, in his famous phrase, “clearly and distinctly”, that it could not be the case that he did not exist yet was thinking. Since he is thinking (specifically, doubting), then in exactly the same way as he can infer that it is true that 2+2=4 from the inconceivability of its not being so, he can infer that he actually does exist from the fact that he is thinking (doubting).

Introspecting into the self the existence of which he has thus established on firm grounds, he discovers that he has in himself certain ideas. These ideas include mountains, animals, etc. But they also include a subset of other notions of perfection, of which the key one is the composite idea of a being more perfect than himself in every conceivable way. He can see, he says in the Discourse on Method, that he is not perfect, since he has in him “the idea of a being more perfect than mine”. This idea, which it is “manifestly impossible” to regard “as nothing”, cannot come from something less perfect than him (perfect, I understand it, in terms of reason). Having thus argued that he exists, has within him the idea of a more perfect being, and that this idea is clearly not nothing, he goes in for the great conclusion:

And because there is no less repugnancy that the more perfect should succeeed from and depend upon the less perfect, than for something to proceed from nothing, I could no more hold it from myself: So as it followed , that it must have been put into me by a nature which was truly more perfect than I, and even which had in it all the perfections whereof I could have an idea; to wit, (to explain myself in one word) God.

Descartes is often said to make use of the ontological argument for God’s existence (originally presented, I believe, by St. Anselm). I found it interesting that actually reading Descartes’ version, it does not really follow the standard summary of the ontological argument, which is: “We can imagine a perfect being. A perfect being would not be perfect if it lacked existence. Hence, a perfect being exists (and we call this being ‘God’).” (A wonderfully cunning argument. I believe that the standard view among philosophers is that it was Kant who finally demolished it by showing that existence is not a predicate in the standard way.) Now, Descartes argument says, instead, that he has the idea of perfection within him, and it could only have originated from something which is perfect. The argument here hinges on the premise that the idea of something perfect necessarily originates from a something perfect in the relevant way, or invertly, that the idea of something perfect cannot originate from something which is not perfect in that way . Hence, since the idea of perfection exists in us, the perfect being from which it must originate, exists (or theoretically, has existed). Importantly, the conclusion that God exists and has the qualitites of perfection is one which Descartes sees on par with mathematical proofs. It is one which no rational being who properly understands the argument can, in good faith, doubt. Thus he says:

As for example, I well perceive, that supposing a triangle, three angles necessarily must be equal to two right ones: but yet nevertheless I saw nothing which assured me that there was a triangle in the world. Whereas returning to examine the idea which I had of a perfect being, I found its existence comprised in it, in the same manner as it was comprised in that of a triangle, where the three angles are equal to two right ones; or in that of a sphere, where all the parts are equally distant from the centre. Or even yet more evidently, and that by consequence, it is at least as certain that God, who is that perfect being, is, or exists, as any demonstration in geometry can be.

Let us get back to the thread of the main story. How does this help Descartes in restoring proper foundations for knowledge, and in particular, scientific knowledge? Well, he goes on to leverage the conclusion that God exists to put back on the table of knowledge all those articles which his method of doubt has temporarily, as part of the skeptical exercise, displaced. If God is perfect in every way imaginable, he is also perfectly good, and so would not deceive the senses so, that the world around us is all the results of the machinations of an evil demon, that what we take to be real is not in any way so. A perfectly good God would not deceive us. Hence, we have reason to trust what careful examination and experimentation can indicate to us about the physical world.

It is indeed amazing how the commonly derided skeptic dreamer Descartes can thus very practically argue:

For they [the notions touching natural philosophy] made it appear to me, that it was possible to attain points of knowledge, which may be very profitable for this life: and that instead of this speculative philosophy which is taught in the Schools, we might find out a practical one /…/ as we know the several trades of our handicrafts, we might in the same manner employ them to all uses to which they are fit, and so become masters and possessors of nature.

For the sake of, he continues, convenience and good health – a most practical view indeed, or at least in comparison with the caricature of Descartes the skeptic. He concludes that he will spend his time studying medicine, as it is the most important science. “I think, therefore, we have solid epistemological foundations for the kind of scientific knowledge which we should use to bring convenience and good health to the public.” Perhaps less catchy, but, if I am not mistaken, a tad more complete a summary.

I always found the idea of realism in political theory somewhat confusing. I am not here talking of realism in the International Relations sense, i.e., as a theoretical outlook stressing the strong systemic or human forces which make international (and sometimes also domestic) politics perpetually brutish and nasty. Likewise, I am not talking of ethical realism, the view that moral propositions have truth values and that some are true, rather than merely hidden expressions of emotions, commands, wishes or other prescriptive judgements. I am instead talking of realism specifically applied to political theory (or philosophy, I make no distinction here). In this text, I first spell out the initial dilemma to which the notion led me. This dilemma suggests realism in political theory is either trivial or in contradiction with the normative aims of the field. I then suggest three senses of realist political theory which avoids both horns of the dilemma, namely: realism as emphasis on institutions rather than abstract theory; realism as a criticism of “utopian method” in political theory; and realism as the affirmation of certain pessimistic empirical assumptions. I of course make no claim to originality here, but hopefully, it can help to clarify certain points about the political realisms available to us.

I won’t try to define political theory here, because any definition might well pre-empt or prejudice a discussion about how it can or could be realist. But let us just identify political theory roughly as the thinking about the normatively best forms of political society and institutions, the nature of and relation between different political values, and the relationship between morality and politics. This suggests that something like John Rawls’ project to construct a just basic structure for a pluralist society, in light of certain conceptions of rationality, dignity, autonomy, and fairness, is a paradigm example of political theory.

What is it, then, that I have found somewhat peculiar about the idea of realist political theory? Let us first look at what the word ‘realist’ suggests. In International Relations, where the term is standardly used, the central point conveyed by a theory describing itself as realist is a certain tough-minded acceptance of the nasty nature of international politics. Realist theory, its proponents say, offers theory grounded in the real facts of the real world, however discomforting these may be. Its polar opposite, supposedly, is naive theorising which fails to account for the realities such as selfishness, fear and cruelty. Its essence can thus be encapsulated in the idea that theory must hard-headedly take into account certain unchangeable realities, in particularly the messy, troubling, complicated realities of human society. This is broad and general, but I believe it captures the way realism has been used in discussion about domestic as well as international politics, though for various reasons it holds a more clearly defined place within the discourse of the latter. This gives us some hints for where to start when we think about realism in political theory.

If, as seems reasonable, we begin thinking of realism as taking into account brute realities of the world, the idea of realist political theory indeed starts to look a little bit peculiar. The difficulty can be put in terms of dilemma, a dilemma between two ways of understanding realist political theory. It is a potential dilemma because both alternatives which I shall describe seem to discredit realist political theory, in different ways. Let me being by summarising the horns of the dilemma. First: If realist political theory merely means political theory which takes the world into account, which political theoretician would not accept that his theory is realist? Who would admit that his theory is completely and utterly detached from the real world? Even the most utopian of thinkers, a Plato or a Kant, will have to say that even if the societies they describe could not plausibly come about, their theorising still “takes the world into account” in important ways. That is, their theories present ideals towards which real humans in the real world better start working, or models from which real, actual people can learn, or conceptual constructions from which we can see what is good and what is bad in the real, dirty world around us. Even if we write about the loftiest, most unreachable of ideals – so long as we admit those ideals are ideals for human beings, and might have looked different had human beings been relevantly different, we will hold that our theory takes realities into account appropriately. Whether a theorist presents down-to-earth concrete changes to voting laws, or elaborates a highly abstract theory of ideal communication, his theory is thought to apply to the real world in some relevant sense. In other words, since no political theorist will admit that his theory does not take the world into account in the appropriate way, how is realist political theory supposed to be different from political theory generally? This is the first horn of the dilemma: the seeming triviality of realism in political theory.

In the above weak sense, ‘realist political theory’ comes to mean simply ‘political theory’ – realist adds nothing to the notion of connectedness to the world inherent in ‘political’. It is now time to see what happens if we exchange this conception of realism (which seemed to lead us to triviality) for a more substantive realism. To avoid triviality, we may try to substantiate the notion of realism to help differentiate realist political theory from political theory simpliciter. If we can specify realist demands on political theory which are stronger than those springing already from ‘political’ per se, we may be able to show that realist political theory is indeed a meaningfully distinct approach to political theorising. Such specification of more strenuous realist demands would, I think, have to take the form of more extensive limits on the “loftiness” of the theory. It may be partly helpful to envision a spectrum, a line. On the lofty, idealist end we may be said to find Plato’s Republic, and slightly further on, the more realist Laws. In the Republic, the major thrust of the project is doubtlessly idealist, in the classical sense. An ideal state is mapped out, the inhabitants and structure of which would be radically different from exiting society (today as well as in Ancient Greece). In the Laws the political change envisioned is not so radical, and the possibilities for a transformation of seemingly intractable human character traits not quite so hopefully expansive. Imagine, then, that pushing the realist criterion further would mean moving from the lofty heights of the Republic, through the Laws, to end up with the realism of the institutional engineer who devotes his energy to arguing the virtues of the Saint-Laguë over the D’Hondt formula for local by-elections. On this simplified image, realist political theory may be characterised as theorising which demands we inhabit a position appropriately distant from the most idealist Platonic loftiness.

However, strengthening ‘realist’ thus, we run up to the second horn of the dilemma, namely, that when pushed too far, realism in political theory seem to deny the normative nature of political theory. The aims of political theory are not merely empirical, but normative and transformative, and if this is so, then why should one accept any more substantive realist demand than the trivial realism-as-relevance we discussed above? If the charge to a theory is that it is not realist enough, the reply could simply be that it deals with what ought to be, rather than with what is (or could easily come about). One may indeed ask why political theory should be more realist than would be entailed by being relevant, by being political. And if we accept that political theory always ought to try to be relevant, what would actually distinguish specifically realist political theory?

The above dilemma encapsulates the confusion I originally felt when encountered the concept of realist political theory. My position was indeed that “Of course it should be ‘realist’ as in ‘relevant’, and of course it should not be ‘realist’ as in ‘limited in its normative scope’. So what does it mean to say that we should try to make political theory more realist?” Seeming dilemmas invite attempts at clarification. In light of the above I will try to suggest how I think we can helpfully understand realist political theory. I say ‘can understand’, because people may of course use the term as they wish. But I think that my suggested understanding has advantages: it specifies intelligible positions and ties them to the historical meaning of the concept of realism. The first point will be one about political theory’s emphasis. The second will be about its method. The third will be one about the empirical assumptions of normative political theories.

My discussion is predicated on the following reasoning: It does not make sense to simply adopt realism as a condition on political theorising which requires it to take reality into account. In isolation, without further context, this understanding of realism is either trivial or unacceptably limiting. But the talk of realism in political theory has been brought forward in response to the way in which political theory has been conducted, so we have reason to at least investigate what impulse is behind it, and if in what sense a dose of realism can improve political theory as practised in the Western philosophical tradition.

The first sense in which I will suggest a political theory can be realist has to do with its methodology and area of focus, or emphasis, rather than the substantive issues involved in specific political theories and problems. In this sense, being a realist political theorist means taking critical stance towards what we may call the utopian tradition in political theory, towards the pre-eminence of ideal over non-ideal theory. It is the kind of criticism launched by Jeremy Waldron when he says that political theorists have had too little to say about political institutions compared to abstract ideals and ideas. The charge is that political theorists talk about the wrong things, rather than necessarily talk wrongly about things. This is not necessarily to say that abstract ideals and ideas have no place, or even a central place, in political theorising. But it is a criticism of just how dominant such abstract approaches or theories have been in the normative discourse about politics. We can understand this as a criticism of emphasis: Political theorists have been too quiet about the worldly applicable, the imperfect but useful, the normativity of engaging with politics as the art of the possible. Whether one agrees with this or not, it is a defensible stance which can meaningfully be distinguished as realist.

The second meaningful sense ‘realist political theory’ can take has to do with method. Here, the criticism is not that political theory has focused disproportionately on the abstract as opposed to practical institutional design and other similarly concrete issues. Rather, the criticism is of the kind of political theorising which proceeds through imagining ideal politics and from such an ideal deduce what ought to be done in the realist’s here and now. It is a moot point here whether many theorists actually get around to working out the real world implications of their idealist theories – it seems to me they do not. But the kind of realism I am conceptualising here is one which criticises this very method, be it actually applied in its entirety or not. We can see this realism at work, I think, in criticisms of Rawls’ veil of ignorance or Habermas’ ideal speech situation. Instead of preceding from a small set of assumptions, through an idealised political situation, towards, possibly, practical advice in a world beset by difficulties not brought in at any previous step in the theorising, one can imagine beginning from assumptions of disorder, complexity, pluralism, etc., and work out theory from this starting point. I guess that, arguably, this is what Rawls tries to do, especially in his later work. The specification of justice as fairness as a political, not metaphysical, idea is supposed to offer a theory which takes as starting point the messy pluralism in modern societies when it comes to theories of the good. It also seems to be what Bernard Williams’ theory tries to do. Williams highlights (“Realism and Moralism in Political Theory” in In the Beginning Was the Deed) that the first question of politics is the Hobbesian question of order:

‘I identify the “first political question in Hobbesian terms as the securing of order, protection, safety, trust, and the conditions of cooperation. It is “first” because solving it is the condition of solving, indeed posing, any others.’

A lot of deep questions are raised by this realist line of criticism: Is there right one method in political theory? Are methods mutually exclusive? In the end, this leads us to ask what we are actually trying to do when we do political theory, a question which is terribly difficult.

Let me know briefly discuss the third intelligible kind of realist political theory. In essence, this realism is brought out by certain key points of empirical disagreement among theorists. The paradigm example of this kind of realism is the standard line against radical, utopian political theories. Such theories, it is charged, rest on empirical assumptions that do not bear the burden the theories put on them. They make unwarranted assumptions about the forms human relationships can take, the sophistication their biology can even under ideal circumstances allow, etc. The historian Richard Pipes’ criticism of communism is a typical example. His charge has two main parts: First, communist utopianism assumes an alterability of human psychological characteristics which is not possible, at least not in any larger community for any long period of time. Second, and related, communist political theorists have held that past societies have lived harmoniously under forms similar to those envisioned by modern communist theory (Richard Pipes, Communism – A History, 1994), in particular, without any recognition of private property. It need not concern us here what the merits of these criticisms are and to whom they may apply. The point is simply that they are a good example of one particular kind of realism in political theory.

It is interesting to note how different realism in this sense is to the senses of realism discussed above, where realism was aimed at the methods or focus of political theorists. Here, the realism is much more like the realism of International Relations. It is a substantive view, labelled realist, but which will only be realist, in this sense, to the extent that we agree with the substantive empirical assertions contained in it. If we compare a caricature modern realist and a caricature modern liberal International Relations theorist, the difference is not that they have different relationships to reality, in the way that ideal and non-ideal political theorists discussed above can be said to have. Rather, it is merely the case that one side of the debate calls its empirical assumptions ‘realist’. Of course, the liberal, who has stronger beliefs in institutions and possibilities of cooperation than the realist, believes that his empirical assumptions are more realistic than the realists. The realist cannot to be said to be more intimately in touch with reality, in the sense of presently existing conditions, unless we want to start making his case. The same is true of realist political theorists who charge that some brand of more radical political theory is unrealistic in the sense that its empirical assumptions about which events are likely to occur or do occur presently or under some specified future conditions. Realist political theory in this final sense, thus, is not bound logically to any particular substantive theoretical content. However, in light of the connotations and history of realism within the political sphere (broadly conceived), it can most fairly, I think, be applied to theories stressing the intransigence of certain, in particular negative, traits about the structures or participants of human politics.

In sum, then, I started by setting up a dilemma which easily emerges (and which certainly emerged for me) as I first started to consider what realist political theory might be. I then moved on to offer three distinct senses – to do with focus, method, and empirics – in which distinctly realist political theory is neither trivial nor in tension with the normative aims of political theory. Hopefully I can move on, in another blog post, to talk of the merits of each kind of realism. For I believe, indeed, that there are some very serious arguments for all three.