Political Science and Political Theory

[Disclaimer: This turned out to be longer than I originally anticipated, so if it turns out to be too much, my apologies.]

What role should should the deliverances of political science play in theorizing about justice? (Of course, this is just a particular instance of a much more general question, viz., what role should the deliverances of any science play in theorizing about anything? I think the answer to this probably depends upon the domain in which the question is asked; that is, the answer probably differs depending on whether one is talking about, e.g., the role of physics in metaphysical theorizing vs. the role of psychology in ethical theorizing. But I’m not sure about this. In any case, I’ll restrict this post to asking about the role of political science in political theorizing.) It seems to me that the answer to this question depends upon how one intends to use the science.

It seems clear to me that political science shouldn’t play the role of telling us what counts as just, i.e., how things ought to be. Such a use would be clearly encounter Hume’s is-ought problem. So, e.g., if political science shows that democracy reduces income inequality, it doesn’t thereby show that democratic constitutions are just; if it shows that communist regimes never last longer than x number of years, it doesn’t thereby show that communism is unjust. Clearly, for the science to tells us what counts as justice, such statements would have to assume the justice of inequality reduction or that just regimes are enduring; but these conclusions don’t follow from any empirical study.

Another role that political science could play is that of indicating what is possible, thereby enabling us to bring our political theorizing in line with the ‘ought implies can’ principle. I.e., the science constrains what individuals and institutions can be obliged to do by telling us what in fact individuals and insitutions are able to do given their limitations. This seems to be something that Hume is up to in `That politics may be reduced to a science’:

So great is the force of laws, and of particular forms of government, and so little dependence have they on the humours and tempers of men, that consequences almost as general and certain may sometimes be deduced from them, as any which the mathematical sciences afford us.

And further:

It may therefore be pronounces as an universal axiom in politics, That an hereditary prince, a nobility without vassals, and a people voting by their representatives, for the best MONARCHY, ARISTOCRACY, and DEMOCRACY. But in order to prove more fully, that politics admit of general truths, which are invariable by the humour or education either of subject or sovereign, it may not be amiss to observe some other principles of this science, which may seem to deserve that character.

(See also `Of the original contract’, wherein Hume argues that just (i.e., legitimate) authority can’t be grounded upon a social contract because such a contract runs counter to human nature.) In other words, there are certain laws that govern the operation of political associations regardless of particular circumstances, and these can be deduced by doing political science. These laws, in turn, constrain which scenarios can count as just. Any theory of justice that posits a situation that contravenes these laws must be discarded. Call this the Constraint View.

The Constraint View seems to be a plausible contender for the role of political science in political theorizing (even if I’ve falsly attributed it to Hume). Kant makes available the resources to construct another candidate role, viz., the role of indicating which provisional steps best serve the end of reaching an ideally just state of affairs. Call this the Provisional View. (Much of what follows is draws on Lisa Ellis‘ argument in Kant’s Politics. Ellis, and Kant, are not concerned with the question I’m considering here, but what they have to say has potentially interesting implications for my question.) This view depends on Kant’s distinction between `provisional right’ and `conclusive right’. Consider, for example:

Prior to a civil constitution [i.e., prior to entering a just state of perpetual peace, wherein each citizen enjoys freedom and equality] (or in abstraction from it) external objects that are mine or yours [i.e., property] must therefore be assumed to be possible, and with them a right to constrain everyone with whom we could have any dealings to enter with us into a constitution in which external objects can be secured as mine or yours. — Possession in anticipation of and preparation for the civil condition, which can be based only on a law of a common will, possession which therefore accords with the possibility of such a condition, is provisionally rightful possession, whereas possession found in an actual civil condition would be conclusive possession. (Rechtslehre 6:256-7)

And again:

In accordance with reason there is only one way that states in relation with one another can leave the lawless condition, which involves nothing but war; it is that, like individual human beings, they give up their savage (lawless) freedom, accomodate themselves to public coercive laws, and so form an (always growing) state of nations (civitas gentium) that would finally encompass all the nations of the earth. But, in accordance with their idea of the right of nations, they do not at all want this, thus rejecting in hypothesi what is correct in thesi; so (if all is not to be lost) in place of the positive idea of a world republic only the negative surrogate of a league that averts war, endures, and always expands can hold back the stream of hostile inclination that shies away from right, though with constant danger of its breaking out. (Toward Perpetual Peace 8:357)

In the first quote, we see Kant distinguishing between provisionally right possession and conclusively right possession. The latter is ideally just possession, but such rights only apply in an ideally just civil condition; the former is less than ideally just, but can neverthless be considered rightful in a less that civil condition. In the second quote, we see Kant distinguishing between an ideally just international state of affairs—viz., a state of nations presided over by a sovereign power—and a provisionally just state of affairs—viz., a cooperative league of nations that works toward peace. The similarity between these two cases is this: the ideally just state of affairs is what is demanded by reason and is the end for which humanity ought to collectively strive, whereas the provisionally just state of affairs, while not ideally just, can be considered just insofar as it advances human society toward the ideally just state of affairs. The relevant Kantian maxim is this (in Ellis’ words): always `preserve the possibility of progress’ (Kant’s Politics, p. 184). Kant clearly recognizes that realizing ideal states of justice may be impossible, but that’s beside the point:

What is incumbent upon us as a duty is rather to act in conformity with the idea of that end, even if there is not the slightest theoretical likelihood that it can be realized, as long as its impossibility cannot be demonstrated either…. [W]e must act as if it [i.e., perpetual peace] is something real, though perhaps it is not; we must work toward establishing perpetual peace and the kind of constitution that seems most conducive to it… (Rechtslehre 6:354)

In other words, ideal justice might be a pipedream, but we are nevertheless obliged to take provisionally just steps toward that end.

On this view, there are no empirical limitations on what counts as an obligation of justice; the only limitations are the dictates of reason. Thus, political science sets no restrictions on what can count as just. It can, however, tell us which steps are provisionally just, that is, which policies best preserve the possibility of achieving ideal justice. The deliverances of political science, then, belong to the realm of non-ideal theorizing—i.e., to theorizing about how to get from an unjust to an ideally just state of affairs—and are excluded from the realm of ideal theorizing—i.e., theorizing about what is ideally just.

At this point, I’m interested in adjudicating between the two views. Hume’s proposal, while more modest, also seems more sane than Kant’s. It appears to be a maxim of common sense that people can’t be obliged to do that which is impossible for them to do. Thus, political science (along with psychology and sociology, among others) determines the constraints within which theorizing about justice can take place. However, there is something appealing about Kant’s proposal as well (which makes me hesitant to concede Hume’s view), viz., it refuses to settle for less than the ideal. Hume’s view could (but need not) lead to defeatism—`This is the most we’re capable of; this is the best we can get’. With Kant, there is no settling for second best; our ultimate end must always be an ideally just state of affairs. Such a demanding theory could lead to despair—`We’ll never be able to meet the demands of morality!’—but Kant guards against this by providing for provisionally just states of affairs. We can always do better, but we need not despair that we can never achieve some degree of justice. Accordingly, political science tells us what is possible in our current circumstances, enabling us to determine what our next provisional steps ought to be. It reveals to us the consequences of certain policies and institutions under our current conditions. With this information, we will be able to better predict which policies and institutions will produce provisionally just states of affairs, gradually moving us closer to the ideal.

Of course, for political science to be able to tell us what counts as provisionally just, we need to know what counts as ideally just, which is itself a matter of heated contention. But that’s a topic for another time.


3 Responses to Political Science and Political Theory

  1. Shen-yi Liao says:

    i am not sure how the kantian proposal works. is it (a) as if from a view from nowhere, we see this progression from the anti-ideal to ideally just, and then we identify where we are, and then see how we can get to the ideal, or (b) we look at where we are now, decide what the ideal is, and see how we can get there from here? it is conceivable that (b) is problematic because what looks like the best option at each stage (i.e. the option that’ll get us to the ideal) may not be the best option in the long run. (i have in mind something like rouble trouble from http://fas-philosophy.rutgers.edu/arntzeni/B&B.pdf but substitute utilities in place of money.) but it seems like (b) is closer to what political science is doing, and in fact, the only conceivable way of doing it. so that looks bad.

  2. Brian Berkey says:

    Perhaps I’m missing something, but the Humean/Constraint View seems to me clearly wrong. Political Science tells us only how individuals and institutions actually tend to behave, not how it is possible for them to behave. We might infer certain things about what is possible on the basis of their actual observed behavior and certain statistical facts about such behavior (as discovered by political science), but then we’re just doing more philosophy, and it’s not clear how conclusions about how it is possible for individuals and institutions to behave can be established on the basis of observations and analysis of their actual behavior. It seems that additional assumptions about, for example, human nature will be necessary in order to reach any conclusions about what is possible.

    The Kantian/Provisional View seems to me more plausible, so long as it avoids positing any limit in principle to how individuals and institutions can behave. Observing how they behave in the actual world is, of course, important in order to determine the best way to move forward toward an ideal, and the view seems to take this into account in a rather plausible way, without giving up the possibility that over time individuals’ and institutions’ dispositions can be altered so that we can continue to progress toward the ideal. The role of political science on this view is, I think, best thought of as purely instrumental. The conclusions reached by empirical political scientists can help us determine how best to pursue progress in circumstances in which people’s dispositions are less than ideal. But unlike on the Humean view the conclusions of political science don’t themselves lead us to draw any philosophical conclusions.

  3. David Wiens says:


    I don’t think (a) is anything Kant had in mind, and it’s certainly not the way I wanted to think about Kant’s proposal. The limitations Kant puts on human reason would prevent us from taking such a bird’s eye view. But, further, his discussion of provisionality seems to take him away from proposing anything like (a); that is, I don’t think Kant thought there was a predetermined course of progress that we, from a view from nowhere (or from the view of human reason) could envision, find our place on that continuum (like locating ourselves with a ‘You Are Here’ dot on a mall map), and then figure out where to go from here. (He does, however, seem to come close to this when he talks about progress being justified by the teleological course of history, something like an invisible hand story. But this teleological progression is blind and is supposed to occur regardless of human choices. So it’s different than (a). But this raises a whole other can of worms.) To be sure, he thought there was an ideal destination to aim for; but the space he leaves for provisional justice gives us room to make ‘false starts’ at reaching that ideal—i.e., we might think something will get us closer to the ideal, but, in the long run, it turns out that it takes us farther away. But this will be ok for Kant; something is provisionally just so long as it leaves open the possibility of long-run achievement of the ideal. For Kant, the long-run view always takes precedence. That’s why he’s willing to tolerate a less-than-ideally-just reality; sometimes pressing toward the ideal too quickly (say, via violent revolt) precludes the possibility of reaching the ideal in the long run, whereas tolerating some injustice in the short run prepares the ground for long-run justice. So, I think your (b) is closer to what Kant had in mind, and I think the best reading of that view overcomes your worry (although my inadequate rendering of it here might not have convinced you yet).


    It’s not clear to me whether Kant thinks there are any principled limits on behaviour. But, in any case, it is clear that any such limits don’t preclude the obligation to pursue ideal justice:

    So the question is no longer whether perpetual peace is something real or a fiction, and whether we are no deceiving ourselves in our theoretical judgment when we assume that it is real. Instead, we must act as if it is something real, though perhaps it is not… And even if the complete realization of this objective always remains a pious wish, still we are certainly not deceiving ourselves in adopting the maxim of working incessantly toward it. For this is our duty… (Rechtslehre 6:354f, emphasis added)

    So, even if Kant thinks there are limits in fact (which, like I said, it isn’t clear to me that he does), such limits don’t prevent him from making it our obligation to take provisional steps toward ideal justice. I think your gloss on the Provisional View—i.e., political science as purely instrumental—is right; at least, it seems to capture what I had in mind when proposing the view.

    However, I would like to push back a little on your criticism of the Constraint (Humean) View. It seems to me to be straightforward empirical science to collect some observations about actual phenomena—say, human behaviour under certain conditions—and then, if our test sample is sufficiently large, infer certain things about the nature of those phenomena—in this case, infer some things about human nature. Isn’t this how science proceeds? (I’ll be the first to admit that this might be a naive picture of science.) In what way does this inferential step from observation of behaviour to conclusions about human nature involve something more than scientific reasoning? (You suggest this inferential step involves philosophy, but it’s not clear to me what you mean by that. It is clear to me why such an inferential step is philosophically interesting, but it’s not clear why the inferential step constitutes ‘doing philosophy’.) I guess I need to hear what you think about these questions before I understand your criticism.

    Pushing this aside, I’m also not clear on why we can establish what’s possible for human behaviour on the basis of generalizations from their actual behaviour. If it’s true that we can draw general conclusions about human nature based on ordinary scientific reasoning (which we haven’t agreed upon), and if it’s true that human nature constrains the range of possibilities for human behaviour (it’s not clear to me how this couldn’t be the case), then political science can help us define the range of possibilities for human behaviour. To be clear, I don’t think that the range of possibilities given here is absolute; as human nature evolves, the range of possibilities might evolve. But political science would deliver the range of possibilities for us as we currently are, which, in turn, constrain what can be considered our obligations to justice.

    Anyway, I’m sure there’s more to say here, but this has, again, grown long enough for now.

%d bloggers like this: