Brendan is right when he says that short pieces make for easier conversation. As such, I apologise for the fact that this post is very long, and just hope it is worth reading to the end.
This post is an attempt at a direct response to a few comments under my piece on the vagueness of rigour. The question was raised as to when precisely Analytic rigour might distort, rather than clarify, what we are attempting to say, or when Analytic precision might be bad precision. I am going to beging by trying to clarify what I mean by ‘ambiguity’ and what I think this might mean for our practising philosophy. I’m then going to attempt to argue that Analytic rigour can have a distorting effect when applied to the interpretation of Scripture in the life of the Church, and that this has potentially serious repercussions for the practise of theology. Specifically, I am going to try and argue that insofar as theology is thought to proceed from Scripture, then it is to some extent premised on a number of ambiguities which are both formal and fruitful, such that they may well not be appropriate subjects for Analytically rigorous approaches.
I’m going to begin by attempting some clear distinctions and semi-formal definitions, lest I be accused off the bat of unrigorous thinking. First, I’m going to distinguish between vagueness and ambiguity. I shall then make a couple of observations about how our understandings of the place of ambiguity in reality can inform different philosophical approaches.
As I am using it, vagueness is akin to blurriness: a vague account of something is thus akin to what we see when we look at something through translucent glass (or, perhaps, a mirror darkly). An example of a vague account might be ‘the game was longish’, or ‘I saw a figure in the fog, but I couldn’t tell if it was a dog or a hedge.’ In a large number of cases, we can clarify vague accounts though further investigation or more exact measurement: so, for example, we could say ‘the game was 3 hours, 4 minutes and 5 seconds long’, or ‘when I checked, it was a hedge’.
I am using ambiguity (uncontroversially) to mean a linguistic utterance which could have two or more distinguishable significances: to take a clear example, ‘Peyton Manning is sick’ could mean either that he is an incredibly good quarterback or that he is feeling ill. To take a less obvious example, ‘she’s a strong person’ is also ambiguous. Not only could ‘strength’ denote a number of different qualities, ‘person’ can also be used in a quietly ambiguous sense (for example, its significance will differ slightly depending on whether we’re talking about a real or a fictional person).
Now, vagueness can often imply ambiguity, and ambiguity can often imply vagueness. The two are not, however, co-extensive. To manipulate a famous example, let’s consider Wittgenstein’s duck-rabbit. We could, in one instance, look at a blurred picture and say ‘I can’t tell if its a duck or a rabbit,’ which would be an example of vagueness. We can also, however, look at a picture composed of entirely clear lines and say ‘it can be both a duck and a rabbit, depending on the aspect under which it is viewed.’ Here the ambiguity is not due to insufficient clarity: it is there in both the drawing and the seeing. This kind of ambiguity I’ll call formal ambiguity, on the basis that the form of the thing referred to is itself ambiguous. (This formal ambiguity can, if it exists at all, exist in vague entities as well.)
I would claim that there are a significant number of formally ambiguous things in the world of philosophy. Among these are things like the duck-rabbit, whose significance one could refer to as aspective (made up word alert?); that is, that what they mean depends upon the aspect under which they are considered, understood broadly as a collation of the external and societal factors relevant to an object’s significance. To take a politically charged example, the swastika is an object whose significance is aspective, dependant on whether we see it on a Nazi flag or the walls of a Norwegian church (and even then, how we see it depends as much on where we have been as on where we see it). Linguistic utterances of any kind, moreover, are also aspective: the significance of a word or gesture is uncontroversially (I think?) contingent upon a wide array of personal and communal factors.
(A key feature of the term ‘aspect’ is that it does not, like the word facet, imply an incomplete perception of its object: we do not perceive part of the duck-rabbit when we see it as a duck only; we perceive the whole thing under a non-totalising aspect. Similarly, when we look at Jesus Christ and see him as completely human, we still see Jesus Christ in his entirety; the point is that Christ’s being completely human does not preclude his being completely God as well.)
A second set of examples includes what could broadly be called family resemblance terms. These are terms which pick out concepts which are not bound together by necessary and sufficient conditions, such that they cannot be unambiguously defined. ‘Game’ is one of Wittgenstein’s most famous examples of such a family resemblance term, and he uses the term ‘language-game’ in part to intimate that ‘meaning’ is another such term (which also implies that ‘ambiguity’ would be one as well). Another example might be the term ‘clear’: we can say that the water is clear, that the coast is clear, that the beach is clear, that our thoughts are clear, that our consciences are clear, that our language is clear, or that its is clear that something is false, and in all of these cases the term ‘clear’ has a variety of implications which, whilst contained within a single concept, are not bound by any unitary thread running through or around them all, such that the concept itself can be considered formally ambiguous. The point is not, of course, that the terms ‘game’ or ‘clear’ are imprecise, but that they precisely pick out such formally ambiguous concepts. (Family resemblance terms can, I think, be vague as well, insofar as they can pick out vague concepts, but I don’t think this is a necessary quality.)
(It is important to note that, under Wittgenstein’s view at least, formally ambiguous concepts cannot always properly be reduced to more logically precise names, like the common example of the many Eskimo words for ‘snow’: to slightly alter a phrase of Derrida’s, what we have with formal ambiguity is ambiguity irreducible to polysemy. It is also worth noting that aspective significance and family resemblance can be related: that it can be the fact that it can be seen under different aspects which renders a specific concept as one bound by family-resemblance (‘blue’ for example, which can be understood as both a colour and a mood depending on the aspect, but which is still a single concept across its various uses).)
Analytic Precision and Its Usefulness
Now, let me give a few examples where I think analytic rigour is extremely helpful when dealing with formally ambiguous terms. It is extremely useful for clarifying specific uses of words so that people engaging in dialogue can perhaps see where they’ve been talking past each other due to an unperceived ambiguity. It is useful in clarifying why we think what we think; to take a crude example, we can perhaps understand someone’s antipathy to religion better when we learn that they think it must be irrational and that they define ‘rational’ as ‘based on empirical evidence’. In line with these examples, Analytic rigour can be very useful for figuring out what people are trying to say and why.
What about cases where Analytic rigour can cloud the issue however? Well, I would claim that Analytic rigour can distort the issue at hand is when its attempts at formal definition lead us to forget the nature of the terms and concepts we are dealing with; when we conflate what we’re trying to say with what we’re actually saying. So, for example, we might offer a formal definition of a family resemblance term, and this might help with communication. If we then take the significance of proper use of the term to be therefore limited by the formal definition, however, we can then illicitly suppose that the term must therefore be used according to that formal definition in all cases; that is to say, we can forget that even though we have offered a formal definition of our particular use of the term, the term itself remains ambiguous in virtue of either its formal ambiguity or the formal ambiguity of its subject.
Similarly, when we offer a formal definition of something with aspective significance, we can forget that the significance of the thing in question is determined by far more than those things that we might pick out as belonging to the essence of the ‘thing itself’: we can ascribe it formal properties on the basis of things which are actually external and contingent, but still essential to the aspect, and so confuse the contingent for the necessary and the intrinsic with the extrinsic, all the while constructing a confused picture of what it is to be ‘essential’ to something’s being what it is.
These are fairly clearly particular variations/interpretations of Wittgenstein’s general diagnosis of philosophical problems (which I am helpfully going to provide no direct quotes for): that problems emerge when we suppose concepts/meanings/language to always behave in the same way, or suppose that they must always behave the same way if they are to be formally significant (I refer to all three of concepts, meaning, and language because I think arguments can be made for all three, but that they are potentially distinct enough to be worth distinguishing between). To put it a bit more pithily: under this view, problems arises when we suppose that conceptual unity must imply uniformity of significance, meaning, or use.
Now, there is obviously nothing intrinsic to Analytic rigour itself which must, of necessity, give rise to these problems, nor does the mooted possibility of these problems furnish a general argument against Analytic rigour; there is nothing wrong with being clear about what we’re trying to say, nor with trying to clarify our statements according to Analytic methods. If, however, our goal is to arrive at formal definitions, rather than just beginning with them, or to get rid of ambiguity, as opposed to clarifying cases where we are using something ambiguously, then I think we’re in trouble; for if we are applying these methods for this reason then we aren’t just using them to be clear about what we’re saying, but to try and limit what we should be saying in a way that might not be true to the subjects of our inquiry.
Why might this actually be a problem? Here’s a list: to do this can also do a disservice to fields of inquiry which do not lend themselves to the use of formal definition (literary criticism, for example), as Frege does in The Thought. On a similar note, an overemphasis on beginning with formal definition can exclude cases of philosophical inquiry where we aren’t clear about what we’re saying at the outset, or where the purpose of our discussion is not to compare the virtues of certain definitions in order to arrive at formal definitions, but instead to explore a language in order to discern how it works. Similarly, it can exclude from consideration the validity of discussions beginning with the assumption that the relevant subjects are formally ambiguous (though such discussions should be clear about this); such an assumption might eventually turn out to be misguided, but it shouldn’t be discounted at the outset on the basis that they aren’t rigorous enough. These methods can also embed their own presuppositions into the debate so firmly at the outset that it biases the inquiry in its favour (by setting the terms of the debate, we immediately handicap those who wish to disagree with them. This is obviously not unique to Analytic philosophy). Finally, it can set dialogue up to fail by positing the ultimate goal as general conformity to a single formal definition, rather than looking at what the interplay of different but hierarchically equal accounts of the same phenomenon have to teach us about its nature.
Now, I would suggest that within the above paragraph there are two relatively distinct conceptions of philosophy coming up against each other. The first is a conception of philosophy as taxonomy; that is, the systematic categorisation of individuals according to specific features (‘specific’ here being used as ‘the defining features of a species’). I think this form of philosophy is broadly similar to the vision outlined in Aristotle’s Metaphysics, with the added proviso that this taxonomy gives us access to the substances of things which are considered as explanatory causes. This is what I have in mind when I refer to a philosophy whose goal is to arrive at formal definitions. The second is philosophy as topography, which is an attempt to chart what we find in the same way as a map-maker. Here, philosophy is cast as a more descriptive pursuit with a less specific goal, and a far less explanatory purpose.
These distinctions should not be considered absolute or absolutely exclusive; indeed, a prior conception of the nature of description or of what there is to be described could lead us to say that a good map of reality will take the form of a general/specific taxonomy (and here is where what I referred as a genuine metaphysical question in my post on rigour comes into play). Furthermore, not only does every map also need a key, and so at least some mutually agreed terms, but there are undoubtedly fields of inquiry where taxonomy is more appropriate than topography (biology, for example, and the provision of clear formal scientific definitions to help us navigate physics).
If, however, one allows for the possibility of formally ambiguous things, it seems to me that one should also allow for subjects of inquiry where over-allegiance to analytic forms of rigour can cloud the issue as opposed to clarifying it. They can do this by leading us to implicitly seek an inappropriate goal, by closing off lines of valid inquiry, by leading us to posit problems in reality where the problem is our over-allegiance to the rigid application of formal definitions in inappropriate contexts, by disparaging other fields of inquiry, and by biasing debate in our favour by embedding our own formal presuppositions into the measures lf correctness employed within the debate itself. None of this is to say that Analytic rigour is intrinsically unrigorous; it is just to say that it is not so intrinsically rigorous as to deserve claim to being definitive of rigour itself (I’m not saying that non-Analytic rigour is rigour itself either: after all, I think rigour to be a vague family resemblance term).
Let me now introduce the second subject of the title of this piece: fruitful ambiguity. As I am looking at it, a fruitful ambiguity is an ambiguity which grants us access to further knowledge, whether by revealing implicit connections between what we’re previously thought to be wholly distinct concepts, belying implicit assumptions contained within certain formulations (personal or communal) and so revealing hitherto unappreciated aspects of the same thing, or laying out new roads which could be taken as we inquire into and practise our philosophies.
True to form, perhaps, I would say that formal ambiguities, so far as they exist, are all fruitful, insofar as they are capable of doing all these things. Despite being coextensive, however, it is still important to use both terms, ‘formal’ and ‘fruitful’, as they point to different aspects of the same thing (see: something like analytic rigour (I hope) in a discussion premised on the usefulness of ambiguity). I also, however, do not want to formally limit what it could mean to be ‘fruitful’: the most comprehensive account I would give in the context of theology is, following Alec’s lead, a vague Scriptural appeal; that a fruitful ambiguity is one which allows us to bear fruits worthy of repentance.
Anyway, to finish this section by giving a far more general, vague, and concise answer to Bill’s question: I would say that analytic rigour is unhelpful when it (intentionally or unintentionally) covers up fruitful ambiguities (this is, of course, a shorter version of the answer given above in terms of formal ambiguities).
Case Study: Scriptural Interpretation and Preaching
Now, Bill and Brendan have both quite rightly remarked that schools of thought should not be generally discounted, even in specific instances, on the basis of their general features. So, let’s look at a particular case study of when Analytic rigour might be unhelpful. This case study will be the interpretation of Scripture as a ground for preaching.
I have to be careful here: after all, to give an interpretation of something often just is to give a formal account of its significance, and (at the risk of revealing some of my more Protestant leanings) I’m going to suggest that preaching is an interpretation of Scripture given so that the Church might hear Scripture speak: as such, preaching could itself be considered the giving of a formal account of the content of Scripture, and so a place where analytic rigour is most welcome. My argument is not about the content of preaching, however: it is about the more general interpretation of Scripture upon which that content is premised. (That said, I think it should be noted that analytic rigour in preaching could be pastorally unhelpful, not least because (as has been noted before) the majority of people seem to have a great level of difficulty understanding what Analytically rigorous accounts of things are actually saying; that, in other words, an analytically rigorous style of preaching could make things very unclear for a congregation (throwing a slightly humorous light on the nature of the aspect of clarity claimed by Analytic thinkers)).
One of the great strengths of Scripture is that it can be variously interpreted by various people in various times and various places in such a way that different interpretations of the same passage can express the truth of that passage to an equal degree. For an empirical demonstration of this, pick one passage in the Gospels, then look up every available sermon ever preached on it (the nativity scene in Matthew is a good starting point). Whilst there are certainly bad sermons which give false interpretations (n.b. variety does not imply universal validity), it is almost always possible to find divergent yet brilliant sermons by preachers from different traditions who offer diverse but profound insights into how these passages might be communicating the Word of God to their present day, insights which cannot be valued higher or lower than each other. (If you want a shorter exercise, then just browse sermons preached on the same lectionary readings in different churches on the same day: the effect is much the same).
Now, I would posit that the reason for this variety in preaching is because Scripture is formally and fruitfully ambiguous. I believe this for both philosophical and doctrinal reasons (for more on which, see my blog post on the secular nature of Biblical literalism), and I on the basis of both of these lines of thought I certainly believe that Scripture doesn’t tend to meet analytic standards of rigour! To take one example (sticking to the Gospels), when the lawyer asked Jesus who his neighbour is, Jesus does not give a formal definition, but a parable about a Samaritan: hardly a paradigm of Analytic rigour, and singularly lacking in explicit formal definition from start to finish. To take another example, Christ never gives a formal account of life, which is somewhat important as it is the grounding concept of two of his most famous proclamations (‘greater love hath no man than this, that he lay down his life for his friends,’ ‘those who seek to save their lives shall lose them’). We could of course look at this as unhelpful and unclear, and believe that we therefore need to supplement Scripture’s vagueness with our own formal accounts. We could also, however, hold that it is actually important for the life of the Church that life is not formally defined at the Scriptural level, such that Scripture allows for a variety of messages to be delivered over the centuries in different sermons. On this basis we might choose to ask with Wittgenstein, ‘who is to say that Scripture really is unclear? Isn’t it possible that it was essential in this case to tell a riddle?’ (Culture and Value, 31).
I would claim that this ambiguity suggests that Analytic rigour is inappropriate at the most fundamental levels of Scriptural interpretation, such that whilst we can certainly seek to interpret Scripture using formal definitions, we must do so in the knowledge that these formal definitions are answerable to Scripture, not vice versa. We can (and probably must) interpret Scripture within the context of a formal apparatus; but this apparatus too must be relativised to Scripture, rather than seen as the means by which Scripture should be expounded. Finally, I think we must sometimes preach Scripture as ambiguous, noting that many of its core subjects are not formally defined, but left fruitfully ambiguous. As such, there are times when our interpretations should indeed eschew Analytic rigour in the name of respect for the object of our inquiry, for the reason that to try and give an analytically rigorous account of Scripture can and sometimes does cover up the fact of its formal ambiguity.
Possible Implications for Philosophical Theology
There is, of course, another central subject of which Scripture resolutely fails to give an analytically rigorous account: namely, God. I cannot think of a point where God’s nature is rigorously defined or placed in the context of a broader formal apparatus. (The Psalms in particular certainly predicate many things of God, but not in a formally defined fashion (and it is well worth remembering that the Psalms are not arguments regarding God’s nature)). Indeed, to pick one point of Christian theology which rest on an inherent ambiguity, the Scriptural use of the term ‘almighty’ with regard to God is barely a distant cousin to the sense in which it is used of Caesar, and importantly so.
I would posit as an empirical fact the claim that Scripture speaks of God using language rife with family resemblance terms and aspective significance; that it rigorously speaks in riddles and precisely denotes subjects beyond the power of our formal definition. I would further posit that this is because, in the words of Barth (and following on from Wittgenstein above), ‘there is a riddle in the fact itself’ (CD, 1/2, 171). I would suggest that this is part of the reason we have the Doctrine of the Trinity, plausibly about as aspective a piece of theology as one can get. I would further suggest that this constitutes the point of miscommunication between Barth and Balthasar on the Analogy of Being (c.f. The Theology of Karl Barth); for even as he treats it as ambiguous, Balthasar still treats ‘being’ as something which can be properly formally defined, whereas Barth implicitly uses it as a term of family resemblance, a fact which has severe consequences for his understanding of analogy,
Above all of this, I would say that insofar as Scripture is formally and fruitfully ambiguous, then our understanding of God and His actions in the world should to some degree allow for that ambiguity in their expression. To this end, I would posit that it is very possible for Analytic rigour to be inappropriately applied to the definition of divine predicates, divine commands, and divine promises, insofar as those predicates, commands, and promises are derived from or found in Scripture.
I am not trying to set up a clear distinction between Analytic and ‘Scriptural’ theology, nor am I trying to say that Analytic theology cannot be Scriptural. I am trying to say that if Christian theology follows, even in part, from Scripture (and in my mind it should in large part follow from Scripture) then it must allow that one of the main sources of our encounters with God is formally ambiguous. In virtue of this, I am trying to say that Christian theology must allow for the possibility that not only can non-Analytic approaches to questions of philosophical theology be rigorous; they can also be appropriate.
On a broader note, I would say that it is (at the least) an open question whether the philosophy of religion should be taxonomic or topographic: for my money, a larger portion of it should be the latter as opposed to the former, and I believe that this is borne out in the Christian religion by the fact that our creeds, rigorously defining as they are, remain unanalysed descriptions of the revealed nature of an un-analysable God. (This is a hypothesis which might be fun to contest.)
A Final Note
This piece has first claimed that there are formally ambiguous subjects of inquiry, then argued that insofar as this is true there are times when Analytic rigour can be inappropriate to philosophy’s subject matter (though it need not necessarily be so). I have attempted to ground a description of two distinct approaches to philosophy (which, whilst not mutually exclusive, do broadly point to different purposes) in my positing of formally ambiguous subjects, and on the basis of this distinction tried to argue that formal ambiguity can also be considered fruitful ambiguity. I have then claimed that Scripture is a formally ambiguous subject, that its formal ambiguity is fruitful for the life of the Church (specifically in its preaching here, though I would argue in other ways as well), and that it is therefore an appropriate subject of non-Analytic rigour. I have then claimed that this has implications for Christian philosophical theology, insofar as it derives at all from Scripture.
I am going to finish by noting that some of the most dangerous formal definitions in history have been, and are, religious. Formal definitions of piety and faith have spurred inquisition and genocide (to invoke Godwin’s law, look at E. Busch’s Karl Barth and the Pietists to see how Pietist self-understanding in the 1920s informed their support of the Nazi party). I would make the case that this is still a danger, and that it is a problem which needs to be analysed and discussed carefully and sensitively. This is certainly not an attempt to make the case that Analytic rigour leads to violence, which would be an offensive, untenable, and opportunistic claim; I do, however, think that over-allegiance to any one form of thought or definition in religious thought, however rigorous it might be, is something we need to be very careful of. Whatever the case, I believe that the discussions we are having here to be important and constructive, even if only for a few, and am honoured to take part in them.