We continue our collaboration with the journal Dao to present featured discussions of a newly published article, available for free download here (link has been fixed). For this edition, Ruth Chang (Rutgers University) has graciously agreed to introduce and share her thoughts about “Comparative Philosophy and the Tertium: Comparing What with What, and in What Respect?” by Ralph Weber (University of Zurich). Ruth Chang’s discussion — and discussion-starter we hope — is here, below. Please feel welcome to join in.
I want to start off by thanking Steve and Manyul for inviting me to jot down a few thoughts about Ralph Weber’s very interesting paper ‘Comparative Philosophy and the Tertium: Comparing What with What, and in What Respect?’ I thoroughly enjoyed the paper and learned much from it.
Weber’s paper begins with a cri de coeur, one with which I couldn’t agree with more: “Comparison is fundamental to the practice and subject-matter of philosophy, but has received scant attention by philosophers.” Most of my research to date explores the nature of comparability in the context of axiology and practical reason, but I have often felt as if I’m whistling in the wind. Understanding the nature of comparison, I believe, is critical for understanding rationality, rational agency, the nature of values, action, and normativity in general. Weber’s paper helps demonstrate how attention to comparison is also important for what’s called ‘comparative philosophy’. It’s gratifying to come across a kindred whistler – especially, since it turns out, we’re whistling much the same tune.
We should distinguish two ‘sorts’ of comparability. One is the sort that ethicists such as myself find most intriguing, what we might call ‘ranking comparability’ – the comparability of items, perhaps reasons or alternatives for choice, with respect to some value or normative criteria that yields a ranking of those items. We might, for instance, rank-compare the act of saving five people as morally better than the act of saving one. Ranking comparability is important for thinking about the grounds of objectively justified choice and action since, it might naturally be thought, you’re morally justified in saving the five over the one in virtue of the fact that saving the give is morally better. Ranking comparability, then, is a precondition for the possibility of justified choice. (This is not to say, of course, that in arriving at a justified choice we must make any comparisons).
Then there is the sort that Weber and probably anyone reading this blog is most interested in, what we might call ‘contrastive’ comparability – the comparability of items where the point of the comparison is not to determine which item is better in a ranking sense but to draw out similarities or differences between items in various respects that help illuminate each or something else to which each or both are related. We might, for example, compare and contrast the Aristotelian and Confucian views of human nature. Contrastive comparability, then, is important as the precondition of a methodology for achieving a deeper understanding of something or of the ways in which two things are similar to or different from one another.
An example of a domain of thought that involves both kinds of comparison is legal adjudication. A judge’s ruling is justified in virtue of being backed by the stronger case (perhaps in light of default rules). And in making her ruling, the judge must engage in contrastive comparison of the present case with prior ones. That’s the analogical reasoning required by the constraint of precedent.
In these short remarks, I want to engage in a bit comparative philosophy – comparing what Weber says about comparison in the context of comparative philosophy, on the one hand, with what I believe holds for comparison in the context of practical or axiological philosophy, on the other. So I’ll be doing a compare and contrast of a view of ranking comparison and of Weber’s view of contrastive comparison. It would be grand if we could unify both sorts of comparison with a single shared structure. Weber’s paper leads the way to thinking that we can.
Weber’s paper defends four main claims:
- Contrastive comparisons require a respect in terms of which they are compared.
- They also require a ‘pre-comparative’ respect in terms of which the things compared can be brought together which may be distinct from the respect in terms of which they are compared.
- Everything is comparable with everything else.
- In doing ‘comparative philosophy’, we should be more reflective about which respects in terms of which we are drawing out similarities and differences so that we can evaluate whether appeal to those respects, such as ‘culture’, are problematic in some way and whether we should explore new respects.
I wholeheartedly endorse 4, about which I will say nothing, though I suspect that is the claim that will be of most interest to readers of this blog. I want to endorse 1 too, and just to break up the monotony of agreement quibble a bit with 2 and 3. They really are just quibbles. My aim is to see if we can maintain a parallel between the structure of contrastive comparisons and ranking ones. I think we can.
As Weber puts it, his leading question is whether a ‘tertium comparationis’ – a third of comparison – is required for contrastive comparison. The tertium comparationis is the respect in terms of which the comparison proceeds.
I’ve argued that a ranking comparison cannot proceed without a respect in terms of which the items are being compared. I call this respect a ‘covering consideration’ and conclude that all ranking comparisons must be three-place relations, such as, x is better than y with respect to the covering consideration, ‘V’. We express an incomplete thought if we say: ‘This bowl of pudding is greater than that shoe’. We need to specify the respect in which it is greater – length? mass? utility as something to throw at someone to show displeasure? I won’t repeat my arguments here because they would take us too far afield. But I’ll continue to talk in terms of a ‘covering consideration’ instead of a ‘tertium comparationis’ for reasons that will become apparent in due course (but also because I made it to only three of my 8 a.m. Latin classes while in grad school…).
Weber’s arguments for the analogous conclusion for contrastive comparison are short, sweet, and to my mind utterly persuasive. Here’s the bottom line:
1) the claim that, e.g., Aristotelian virtue is similar to Confucian virtue, presupposes some specification, if only implicit, of the respect in terms of which they are similar;
2) the claim that those views of virtue share a family resemblance presupposes a variety of points of commonality and so a variety of respects in which they are similar; and
3) the claim that one view of virtue is analogous to the other proceeds either by appeal to a general principle under which the respects of analogy are subsumed, and so rely on a point of commonality for comparison, or, a la David Wong, by case-by-case comparison of ‘sufficient similarity’, thus also requiring a third of comparison.
Thus when we compare and contrast two traditions, we need some respect in terms of which we draw out similarities and differences. Claims of similarity, family resemblance, and analogy all assert claims of commonality and thus rely on a covering consideration.
This all seems right to me. So far, so good. The parallel b/t contrastive and ranking comparability is holding so far.
Weber then goes on to say (moving to 2.) that “it is a mistake to locate the third of comparison only in the tertium comparationis as that in which respect the comparata are to be compared. There is a tertium already required to determine the comparata, and insofar as that determination precedes the comparison that tertium may be thought of as ‘precomparative’. (154) The idea here is that even before we select two comparata we must already have fixed a pre-comparative covering consideration in order for it make sense to bring those very comparata together. For example, we could take the Mengzi and the Xunzi to be two different schools of Confucianism, or we could take them to be two expressions of a single Chinese philosophical tradition (in contrast to a Western tradition). This pre-comparative covering consideration “predetermine[s] whether or not a comparison in respect of the notion of human nature will even be among the considered options.” (163). Crucially, it needn’t be the same covering consideration in terms of which the contrastive comparison proceeds.
This idea of a pre-comparative covering value, however, seems a little strange. Do we really need it? Can we just do all the work that Weber wants done by appeal to a comparative covering consideration – a tertium comparationis? I’ve argued in the case of ranking comparisons that we can. Instead of proliferating respects in terms of which a comparison proceeds – first a pre-comparative respect that allows us to bring together two items, and then a comparative respect in terms of which we compare them – we can instead simplify the domain by distinguishing the ways in which comparison might fail.
In the case of ranking comparability, sometimes items being compared with respect to a covering value are ‘noncomparable’ rather than comparable or incomparable. Two items are noncomparable with respect to a covering consideration if they don’t fall within the domain of application of the covering consideration. (Strictly, the point should be put in terms of predicates, but we can be loose without loss here). Which is better, the number four or a table lamp with respect to justice? Noncomparability holds when the items being compared with respect to a covering consideration don’t fall within the domain of application of that covering consideration. ‘Justice’ doesn’t have within its domain of application abstract numbers or table lamps; that is, the covering consideration, ‘justice’ doesn’t ‘cover’ the items being compared.
Noncomparability isn’t important for ranking comparability because practical reason doesn’t present us with practical puzzles concerning noncomparables. Similarly, it seems to me, noncomparability isn’t important for contrastive comparability because there is nothing illuminating to be gained by trying to compare and contrast two traditions with respect to a covering consideration that fails to ‘cover’ them. So instead of appealing to a “pre-comparative” covering consideration before we even attempt to compare and contrast two traditions, all we need to do is to fix on the covering consideration that matters to us in our contrastive inquiry – what’s the respect in terms of which we want to compare and contrast two traditions? – and then proceed from there. If we are conceptually confused, we’ll get it wrong and attempt to compare noncomparables. But we don’t have to appeal to a pre-comparative tertium to discover this. We can discover this conceptually, by realizing that the covering consideration of interest doesn’t ‘cover’ the items we are trying to compare. (There is an issue about whether the covering consideration needs to cover only one or all of the items being compared, but I leave that aside).
Folding in the work done by a pre-comparative tertium into the work done by the covering value just makes things simpler and allows for a nice parallel with ranking comparability, for which a pre-comparative tertium is otiose. This isn’t a real criticism of Weber but just a suggestion for how to allow the tertium comparationis do all the required work while maintaining a nice parallel with ranking comparisons.
I’ll end with a quick, again small, remark about Weber’s claim 3., viz., that everything is comparable with everything else. This claim is, I think, strictly an equivocation on the term ‘comparable’ Weber uses earlier in the paper. What Weber shows, convincingly to my mind, in the earlier part of the paper is that comparability is a relativized notion, that is, that it makes no sense to talk of one thing being comparable fullstop with another. Comparability only makes sense relative to a specific covering consideration, a respect in terms of which the items are compared. In ranking comparability, I’ve argued, it makes no sense to talk of one thing being comparable, fullstop with another. The upshot is that our very notion of comparability is three-place; it’s not that x is comparable with y, but that x is comparable with y with respect to V. Comparability is itself three-placed. So if we say ‘everything is comparable with everything else’, there is no sensible notion of ‘comparable’ we are using.
Of course, what Weber means to say is that between any two items, there is always some covering value in terms of which they can be compared. I’ve argued that the same goes for ranking comparability. You can compare the number four and a lamp with respect to any instrumental – what Weber calls ‘external’ – covering consideration, such as ‘the contemplation of which brings me greater pleasure’. (I can report that with respect to this covering consideration, the number four is better than a lamp.) Similarly, Weber notes that any two comparata in comparative philosophy can be related by being ‘of interest to me’. As he points out, in contrastive comparison, determining the right covering consideration is of the utmost importance. That takes a kind of philosophical judgment or ‘nose’ for ferreting out the respect in terms of which comparing and contrasting two traditions or theories might yield the most illumination. We shouldn’t, Weber cautions, simply assume that ‘culture’ is always the best tertium over other possibilities, like “gender, age, profession, class…philosophical position…coloniz[ed]” and so on. (pp. 167-8).
So, in sum. There is common, unified structure of comparisons in the sense of ‘ranking’ comparisons – this is better than that – and of comparisons in the sense of ‘contrastive’ comparisons – this is similar to that in such-and-such ways. Both require a covering consideration that ‘covers’ the items being compared. Weber’s excellent article helps illustrate this structure with careful and persuasive arguments.
Ruth Chang, Rutgers University