Most of the discussions I’ve been involved in on this blog have centred around the applicability of different methodologies, each of which carry with them different conceptions of rigour and precision (or, in other words, what it is to be good philosophical writing). The discussion has been immensely fruitful for me, and has (I think) laid clear where some of the clear points of miscommunication between these approaches to philosophy lie. So, I thought I’d write about them. I’m not, of course, necessarily the most qualified person to do so, so be aware that I take everything I say with a pinch of salt (even if I sound very sure of myself whilst saying it).
I’m going to begin by identifying the different approaches of which I’m speaking. I’ll then claim that they can be characterised by their acceptance and rejection of a distinction made by Frege regarding language and truth. I shall then look and how their acceptance or rejection of this distinction impacts first the ways in which they construct arguments and the ways in which they critically evaluate them. My hope is that, by the end, it will be clear that both approaches are deserving of respect from each other in virtue of the perspectives they bring, and that they should as such seek to engage in dialogue from their differing positions in order to help keep each other honest.
Continental/Analytic, Apples and Oranges
The most obvious divide in modern Western philosophy is between the Continental and Analytic schools of thought. I am not, however, going to be speaking directly about these schools. I am first and foremost concerned with the divisions between two specific approaches, and the Continental and Analytic schools are both too diverse to be essentially identified by allegiance to a single approach to philosophy.
In the name of precision, then, I’m going to refer instead to Apple philosophers and to Orange philosophers. Broadly speaking, Apple philosophers are considered to be those who value strict logical accounts of arguments and who consider the highest marks of rigour to include formal definition, distinctions premised upon these formal definitions, and a formal apparatus into which these definitions can be placed. They tend to want to state things unambiguously, for various reasons both ideological and practical. Orange philosophers, meanwhile, tend to try and employ holistic accounts of linguistic utterances in their writings, including many facets of an utterance that Apple philosophers consider to be inessential. They rarely consider vagueness or ambiguity in language to be necessarily bad things (indeed, many of their writings proceed from the ambiguities which can be perceived in the giving of a holistic account of a linguistic utterance).
These brief accounts are not intended to be formal definitions (even within the confines of these made up words there is room for difference): they just seek to give enough of the characteristic features of Apple and Orange philosophers for most readers to have a clear idea what we’re talking about.
A Characteristic Divide
My belief is that the division between Apple and Orange philosophers can be characterised in terms of their acceptance or rejection (implicit or explicit, and allowing for differing degrees) of a distinction made by Frege in The Thought. Specifically, I believe that Apple philosophers (implicitly or explicitly, and to differing degrees) suppose thoughts (in Frege’s sense) to exist, whereas Orange philosophers do not. I believe that this disagreement informs the matter of their disagreements elsewhere.
Let’s first look at Frege’s account of a thought: ‘Without wishing to give a definition, I call a thought something for which the question of truth arises. So I ascribe what is false to a thought just as much as what is true. So I can say the thought is the sense of the sentence without wishing to say as well that the sense of every sentence is a thought. The thought, in itself immaterial, clothes itself in the material garment of a sentence and thereby becomes comprehensible to us. We say a sentence expresses a thought.’
Now, the truth of which the question arises for a thought is, as well as being ‘unique and undefinable’, the defining telos of logic. That is to say, the goal of logic (according to Frege) is to discover the laws of that truth of which the question arises for thoughts. This fits win the notion that the thought is constituted by the logical properties of an utterance: for it is these properties which will determine its function within a broader logical structure.
The fact that truth here is the telos of logic means a strict distinction between the laws of logic and the psychological laws of assertion. This is because ‘the assertion both of what is false and of what is true takes place in accordance with psychological laws[, but] a derivation from these and an explanation of a mental process that terminates in an assertion can never take the place of a proof of what is asserted.’ The laws of truth, however, are precisely those laws which enable us to prove what we assert, and the task of logic is ‘the task of discovering the laws of truth’; as such, those psychological facts which condition the conventional content of our assertions are not the concern of logic. Rather, logic is concerned with the thought behind the convention; the immaterial thing which has taken on material clothing inessential to the question of truth; the logical properties of the expression which bely its essential aspect.
As such, I understand Frege’s ‘thought’ to be something like the logical content of a statement, understood against the paradigm of the logical content of a mathematical utterance: for example, the number ‘5’ has certain formal properties which determine its relationships to other numbers and how it will function in a calculus, and these formal properties can be defined in the terms of that calculus. Frege’s thought (sic.) seems to me to be that it is possible to isolate the logical properties of a common linguistic utterance according to similar methods, that the isolation of these properties allows us to proffer formal definitions, and that these formal definitions can express the essential properties of that thought. These essential properties can then be used to show how a thought behaves according to the laws of truth.
Now, the the laws of truth of which Frege speaks are those (by turns simple, by turns complicated) laws of logic we might learn over the course of our lives; those which determine the truth functions of conditionals, disjunctions, conjunctions, and negations (among other things). These laws can then be used to derive laws of validity, according to which we can know that ‘if P then Q, P, therefore Q’ is a valid argument, whereas ‘if P then Q, Q, therefore P’ is an invalid argument. When it comes to the importance of these laws, Apple and Orange philosophers can (and, I think should) agree.
Where they disagree, however, is the filling in of these skeletal laws; in the fleshing out of these Ps and Qs. I think it fair to say that Apple philosophers generally seek to understand the sentences which will fill out these laws according to their logical content, as understood above: this logical content can then ensure that the sentences behave in a logically consistent way, much the same as the number ‘5’.
It is characteristic of Orange philosophers, however, to believe that the positing of this isolable, essential logical content is at root nothing more nor less than the ideological positing of a phantasm: that is to say, that ‘the thought’ is here an ad hoc theoretical posit motivated by a particular aesthetic sense of what truth should be, and that it is in fact impossible to logically distinguish the logical properties of an utterance from it psychological properties (in the sense of logical employed by Frege). As such, they tend not to attempt to give a formal definition of a term in such a way that we might think that the logical behaviour of that term could be reliably determined or predicted by the properties picked out by that definition. They also tend to focus on precisely those aspects of an utterance which seem to Apple philosophers to be inessential to the logical behaviour of an utterance; on the material clothing whose contingency and insufficiency might lead Apple philosophers to suppose it to be irrelevant to questions of truth (after all, ‘when it is a question of truth possibility is not enough’, and ‘what is only half true is untrue. Truth cannot tolerate a more or less.’)
I’m going to try and clarify a particular way in which this distinction might be understood, based on a session I recently attended about the making of the St. John’s Illuminated Bible. Illuminated Bibles are Bibles in which words and passages have been supplemented by particular drawings which seek to draw out some of their particular aspects, and I think that they can serve as an excellent visual illustration of the difference between Apple and Orange philosophers.
Let’s focus on an Apple philosopher first: taking two different illuminated Bibles, I think that it would be characteristic of an Apple philosopher to look for what the two have in common and then reproduce all that without the illustrations, presenting it as what is essential to the two Bibles. They might then take these commonalities and generalise them as the definitive characteristics of a Bible, until a Bible with different characteristics forces a redefinition.
The Orange philosopher, however, will likely differ in three ways: firstly, they will likely focus on how the different illuminations could suggest that what the Bibles have in common might not be so common as we might think in the absence of explicit illumination. They would as such seek to disrupt the sense in which these commonalities are essential to the Bibles. Secondly, they will likely say that even the Bible produced by the Apple philosophers is illuminated: that the absence of images and the sparseness of the layout carries with it as many implications essential for the understanding of the text in this iteration as in the more explicitly illuminated Bibles. Thirdly, following on from and in virtue of this, they will be more likely to focus on what is essential to a particular iteration that what appears to be essential across iterations: rather than speaking of what is essential to ‘the Bible’, they will be more likely to speak of what is essential to the presuppositions of this or that Bible in its particular presentation, and how these essential aspects influence our understanding of this particular iteration.
Why make this distinction? Because these different approaches to illuminated Bibles can be generalised to arguments: that is to say, Orange philosophy tends to read an argument as essentially illuminated, such that this illumination (though contingent) is essential to its logical character (this is, moreover, especially true of arguments which claim to be free of inessential aspects). And this has several important consequences.
The first thing to note in virtue of this is that Orange philosophers are, in at least one sense, radical particularists: over and above than the general form of an argument, they are interested in the integrity of a particular iteration of an argument in the light of its illuminating/illuminated features (this radical particularism is of course derived from a general comment about arguments, expressed here as the statement that arguments are always illuminated). The second is that, in the light of this particularism, the character of these illuminations is considered relevant to the logical integrity of the argument itself (in much the same way that the tone of the illustrations of a Bible is relevant to its integrity as an iteration of Scripture: we would probably, after all, question an illuminated Bible which showed Christ living the worldly life of a hedonistic millionaire, or one which illuminated the Sermon on the Mount by showing him burning the meek in hellfire). The third thing to note is that, because these illuminations are often emblematic of the psychologically determined content of the argument, such that psychologically determined content becomes relevant to the logical integrity of the argument, the strict division between the logical and psychological content of the argument is broken down. There is thus no ‘thought’ in Frege’s sense; there is nothing to which the question of truth (in Frege’s sense) exclusively applies apart from the laws of psychology and the rules of convention.
The Apple philosopher, meanwhile, is more likely to try and find the general logical features shared by the arguments and evaluate them in virtue of these. These general features may centre around the logical form of the argument, and so its structural validity or invalidity, but they may also be focussed on the logical properties of the terms employed in the argument, and so on the very aspects of the argument which the Orange philosopher believes to be conditioned by its illumination.
It is not my purpose here to argue for one approach over the other. In the terminology I’m employing, I tend to think of myself as belonging to the Orange camp; as far as I’m concerned, however, this makes it all the more important that I engage with others who will be able to tell me where I’m missing or glossing over important general or logical features, when I am not speaking clearly enough to be understood, or where my artificial use of terms like ‘Apple’ and ‘Orange’ gets boring (not that this is a practise specific to any particular school of philosophy!). My purpose is instead to try and clearly lay out how this first divide might influence the ways in which we approach the construction and criticism of philosophical statements, so that we can appreciate what each of us is saying and why, listen, and then mutually inform each others’ views so as to move forward as reliably and conscientiously as possible.
The Construction of Philosophical Statements
I am first going to look at how the different presuppositions laid out here impact the way that Apple and Orange philosophers might go about constructing their philosophical statements.
The Apple approach is, I think, closer to Frege’s, in that it is primarily interested in the production of logically valid arguments which can guarantee the correctness of a particular line of thought. To this end, it tends to be characterised by a formalism which seeks to guarantee, to greater and lesser extents, the logical good behaviour of its terms. This includes the formal definition of its terms, so that their identity across iterations within their argument can be underwritten and so that the relevant logical properties of the terms can be distinguished. They then try to construct a logical argument which provides the strongest possible logical case for holding a particular view, typically ending in a specific conclusion.
It is, I think, important to note that this practise does not require an uncritical submission to Frege’s thoughts about thought: you don’t have to believe that we can wholly isolate the logical properties of a term to think it worthwhile trying to construct clear logical arguments in this fashion. It does, however, share certain characteristic preoccupations with Frege’s writing: it does attempts to distinguish the essential and inessential properties of the relevant utterances in terms of their logical significance. It aims at being as good a guarantor of truth as possible, and so provide as good a guide as possible for what we should think. And in this, I think the Apple approach is primarily ethical (and I think this touches on a post that Alec wrote a while back): it attempts to lay out what we should think in a logically grounded fashion. This, I think, underwrites the ethical significance of Alec’s calls to rigour: that it can seem to be epistemically irresponsible to construct arguments in any other way.
The Orange approach is a little less concerned with providing logical arguments as guarantors for truth. In breaking down the distinction between the logical and psychological properties of an argument, it has a much broader (and vaguer) sense of what is essential to a philosophical utterance (this broader sense is, in and of itself, allowed in Frege’s writing, as ‘what is essential depends on one’s purpose’. It is merely subordinated within Frege’s thought, insofar as he considers the purpose of logic to be a higher calling in the field of philosophy (which, after all, it may well be)).
From my perspective (and I do write more personally here), the thought process runs as follows: arguments can indeed be represented in logical form, and their validity can be determined by ana analysis of the logical form of an argument. The meanings of the words which give a particular argument its content, however, are not determined logically: they are conventional and carry with them many and various different psychological, practical, and social implications and connections (furthermore, I would argue, the logical significance of a term is determined by these factors). Because, moreover, even when we are dealing with a valid argument, it is the content of the argument which convinces us of its truth, that is to say, the words themselves, it is the psychological aspects of an argument which compel or convince, over and above the purely logical; this is why Wittgenstein writes that ‘we are at most under a psychological, not a logical, compulsion’.
As such, Orange philosophy tends to examine and describe the various psychological and conventional aspects of its terms before, and sometimes even instead of, presenting its arguments. Because these aspects are not strictly formal, moreover, attempts to formally define these terms can be seen as counter-productive to this effort: they may lock us into a particular usage which fails to take account of how we are influenced by non-formal factors, an influence which may well become important as we continue the work of philosophical investigation into the truth of utterances or the soundness and implications of arguments. As such it seeks to build philosophical statements by focussing on their illuminations, whether these are constituted by certain family resemblances, their belonging to a broader chain of differénce (which in some ways amounts to the same thing), the particular character of their use as an event, their genealogy across the history of human language, or even something as simple as tone. These illuminations constitute both contingent and essential features of any particular utterance: their contingency suggests that their behaviour as terms of a logical argument might be essentially illogical, whilst their being essential entails that they cannot be analysed away in the hope of leaving a bare and substantive residue. This is a part of the reason why Orange philosophy is so unconcerned about ambiguity: after all, if it’s there then it might well have something important to tell us about what we believe and why, or allow us a more precise expression of an illuminated phenomenon than we would get with a supposedly unambiguous formal defintion.
A part of the governing purpose of this approach is to examine why and how we believe what we believe and what the extra-formal implications and presuppositions of our beliefs might be. As such, it tends to be descriptive before it is prescriptive/ethical. Description does, however, carry an ethical component: after all, the why and how of what we believe is instructive as to whether or not we should in fact believe what we believe in the way that we believe it. The potential implications of illumination, meanwhile, point us towards matters of responsibility which might not necessarily be considered in a more rigorously formal approach (n.b. that ‘rigorously formal’ is here a compound expression: that I am not essentially delimiting rigour in terms of formalism). To this end, descriptions of illumination can play an argumentative role.
Insofar as all this is true to practises that exist, I think it fairly clear where certain miscommunications arise. Apple philosophers can easily think that Orange philosophers are talking nonsense or dealing in frivolities, whereas Orange philosophers can suppose Apples to have far too narrow a sense of what is relevant to the question of what we believe, why we believe it, and whether we should believe it. Apple philosophers can take Orange philosophers to be unconcerned with logic or the proper logical behaviour of concepts, whilst Orange philosophers can hold Apples to place too much faith in our ability to contain concepts by means of formal definitions.
There are also, however, ways in which these approaches can inform each other (even if they insist on supposing that the other is fundamentally mistaken!). When it comes to the construction of an argument, Apple philosophers can remind Oranges of the importance of justification and grounding, as well as pointing out that within our philosophy we should try to say things which are true in ways that others can understand. Orange philosophers, meanwhile, can remind Apples that there is more to truth than logic, and that we should not allow efforts to say things in the most reliable way possible to trick us into offering theories which are not true to the actual functioning of human beings (not letting the ideal crowd out the actual in our thinking). Both can also help to point out blind spots in the other’s thinking.
(It is no coincidence, I think, that the practise of Orange philosophy has far more in common with literary criticism than mathematical analysis, whilst Apple philosophy is more like mathematics than literary criticism: the fact that philosophy seems to occupy a middle space between mathematics and poetry seems to me to be both one of its greatest strengths, as it invites the methods and resources of both, and one of its greatest dangers, as there is every chance that mathematicians and critics will just talk past each other!)
Deconstructing Philosophical Statements
The attitudes of these different schools of thought to constructing philosophical statements inform the means by which they criticise them too.
Starting with Apple philosophers: they will tend, in my experience, to try and either show that an argument is invalid (usually by providing counter examples), that it leads to a contradiction, or that the terms employed are too vague to justify the supposed strength of a conclusion (or that they are are just simply misused). For the most part, I think these tend to be all or nothing practises: a valid counter-example to a particular argument is enough to destroy the argument, or at least force a major reshuffle.
Orange philosophers, however, are far more concerned with what we might call integrity than validity. They tend to try and draw out the implicit, unconscious, or unintended implications or presuppositions of a particular utterance in order to see whether or not these actually fit with its explicit content: the contradiction that arises might not then be strictly logical, but instead a sort of contradiction similar to that of body language. They will tend to take a formal definition and, instead of seeking to show that it is not rigorous enough or that the argument structure within which it is placed is invalid, try to show how its illumination in particular cases undermines its strength within generalised logical arguments, or how its extrinsic qualities impinge on and disrupt our understanding of its formal properties. Contradictions tend to be far less important to them, primarily because their is rarely a presupposition that the fleshed out sides of a statement ‘P and not-P’ will actually behave in the same logical fashion. Similarly, counter-examples play far less of a role in Orange philosophy: this is in part because there tends to be little expectation that a generalised logical argument could ever cover all cases, due to the fact that the terms employed will necessarily be badly behaved (even in general claims like this, meanwhile, counter-examples tend to be glossed over: after all, the mere fact that a term can be function in a perfectly logical way today is no guarantee that it will continue to function in that perfectly logical way tomorrow).
Again, the opportunities for miscommunication are obvious: an Apple philosopher can see Orange philosophy as riddled with inconsistencies whilst their criticisms of arguments completely fail to touch on its essential features. Orange philosophers, meanwhile, can see Apple philosophy as so focussed on a few details that it completely fails to recognise glaring inconsistencies across the whole of its argumentative structure. Thus Apple philosophers can be cast as narrow minded and over-confident, whereas Oranges can be cast as hopelessly vague and entirely fatuous.
There are, however, again important points of dialogue to be built on. It is important for Orange philosophers to be reminded that arguments do have logical structures, and that we must to some extent be responsible to these structures. There are arguments in Derrida and the later Wittgenstein, and though it would be a tragic mistake to try and read their writing in such a way that we attempt to isolate the ‘essential’ features of these arguments, or try and formally define their terms in a way that they explicitly rule out, it would be a similar mistake to pretend that they are not there or shouldn’t be subjected to logical analysis. Furthermore, logic and definition are useful tools not just for suggesting how we should think, but can also show us what we do think, whether we know it or not: as such, they are relevant to the practise of Orange philosophy.
On the other side, meanwhile, it is important for Apple philosophers to remember that there is more to life and truth than logic, and that extra-logical factors can be relevant to the soundness of a logically valid argument. Similarly, it is important to remember that the mere formal definition of a term does not actually limit what that term can mean across the course of an argument: that something so seemingly innocuous as tone can change the meaning of the term ‘good’ over the course of a lecture, in such a way as to render a formally valid argument psychologically unsound. Orange philosophy can as such qualify Apple philosophy, from time to time reminding it of its limitations, from time to time pointing out potential pitfalls which are not covered by the methods of logical analysis.
If you’ve got this far, thank you. And if you just skipped ahead to the end, all is well as I’m going to try and briefly sum up.
The overall claim of this piece is this: that there are two distinct schools of philosophy which can be characterised by different responses to a distinction made by Frege regarding the content of a logical proposition (this difference need not be absolute: you don’t need to uncritically accept Frege’s distinction to be one kind of philosopher, you don’t need to unconditionally reject it to be another). Our responses to this distinction will colour our attitudes to what is important in a linguistic utterance, and through this our attitudes to what is important in the analysis of logical arguments fleshed out in the material clothing of human language. These attitudes will further influence our thoughts on what the purpose of philosophy might be, which further impacts the approaches we take to arguments and utterances. The differences between these two schools of thought are real, but they need not be destructive: from their different vantage points it is possible for their different approaches to inform and point things out to each other. They can engage in dialogue, not with the intention of changing each other’s minds, but in the hope that by listening to each other they might learn something of the truth which they would have glossed over if they had been left to their own devices. Finally, through this dialogue we can build both a more precise and a more accurate picture of the world in which we live; the hope is that we might then be better placed to actually live in this world wisely (which is, I think, the point).
I have called these two schools of thought Apple and Orange philosophy. I have done so because even though I think one roughly corresponds with the Analytic School of philosophy and the other with the Continental School, I am not confident enough that what I have described above could be responsibly generalised across those entire schools (indeed, it may be that the schools of thought I’ve described are just the inventions of my own mind). Whatever the case, I do hope that Analytic and non-Analytic philosophers are able to listen to each other at some point, and I hope that this listening might prove helpful to both.