A view from social psychology
Just a heads-up. This article is dense and semi-academic. But you’re not likely to read anything like it in the mass media or even academic journals, and I think the topic needs more attention and discussion. Popular culture is increasingly influenced by ideas from academic philosophy (or its close cousins, literary, critical, cultural, gender or social theory). But philosophy, or any purely discursive discipline, is not a method for producing good… or even true… ideas, but rather the opposite. If you’re interested in why, read on.
Two English-speaking anthropologists visit an insular tribe that is purported to have its own unique language. In order to learn the tribal language, one anthropologist joins the tribe. The other camps outside, but finds she is unable to learn anything about the tribe’s ‘language’ (other than what it sounds like) by observing their behavior, which is entirely visible and audible to her through surveillance equipment. The tribe members appear to ‘talk’ a great deal, but nothing they ‘say’ seems to relate to the external world. They only ‘talk’ when sitting around doing nothing else, and go about other activities silently (using only crude gestures such as pointing). Some years pass, and the two anthropologists compare notes.
The anthropologist who lived with the tribe now claims to be fluent in its ‘language’, and offers to translate it for his colleague. However, he is unable to demonstrate how his ‘translation’ relates to the tribe’s activities when they are not merely ‘talking’. Applying the standard empirical criteria for the presence of linguistic communication, his colleague concludes there is no way to ascertain if her fluent partner has really mastered a new language, or has instead learned to utter a pattern of meaningless sounds. The ‘fluent’ anthropologist is convinced that his newly acquired language ‘makes sense’; but is unable to demonstrate (even minimally) that it does. There is no way to independently verify the accuracy of any translation he offers.
The non-fluent anthropologist refuses to conclude that her colleague has learned a language, because she was unable to observe tribe members using the ‘language’ to mutually coordinate their physical activities (other than making sounds to one another). The tribal ‘language’ is strongly non-instrumental, in the sense that there are no criteria by which its meanings can be discerned apart from the subjective intuitions of the language-users. As such, there is no way to ascertain if an expression means the same (in other words, is semantically consistent) from one instance of use to another. In that sense, the purported expressions of the tribal ‘language’ are semantically indeterminate.
Language consists of semantics (the literal meanings of words) and pragmatics (what we do with words beyond conveying literal meanings, such as using pitch and volume to create a mood). In the case of the tribe’s language, there is no apparent semantics. The ‘language’ is not observed to refer to anything. Even if the non-fluent anthropologist were to master it via immersion in the tribe like her ‘fluent’ colleague, she would not be able to substantiate if the ‘language’ makes sense (even to herself), or if she has simply learned to make noises that the tribe finds agreeable. Perhaps the tribe does not view its oral noise-making as a language, but rather a form of music (like birdsong), which was misinterpreted by the ‘fluent’ anthropologist as linguistic communication.
The parallels between the tribe’s ‘discourse’ and that of philosophy are striking. The latter is concerned primarily with abstractions and is highly self-referential, eschewing any form of instrumental language use; such as the conduct of real-world (as opposed to ‘thought’) experiments, or technical instruction beyond the manipulation of apparent symbols. In the course of their work, philosophers do not use language to mutually coordinate their interactions with the non-textual physical world. They simply ‘talk’ to one another, like the tribes-people in the example.
For the purpose of this article, ‘philosophy’ is defined as a purely discursive extended inquiry. A discursive inquiry proceeds from theoretical reflection upon general ‘armchair’ beliefs; ones that the average adult person has acquired in the course of a normal life, rather than through empirical observations specifically carried out, or consulted, for the purpose of the inquiry. Not all that is generally called ‘philosophy’ falls under that definition, but certainly enough to make this critique worthwhile. On the other hand, not all purely discursive disciplines are always called ‘philosophy’ (for example, parts of social, critical or literary theory).
Drawing on the ‘two anthropologists’ scenario above, ‘discursive’ in this context stands in contrast to ‘instrumental’ discourse, in which language is used by speakers to mutually coordinate their collective manipulation of spatio-temporal phenomena. Examples of instrumental discourse include builders discussing how to construct a house, a physicist describing Newton’s laws, or a patient explaining her symptoms to a doctor (the patient’s body being the object of manipulation in this case. Psychotherapy is a grey area, more on that below).
There are debates as to whether mathematics or formal logic are discursive or instrumental, but that question is not relevant to the topic at hand, since the bulk of philosophical discourse consists of natural language rather than mathematical or logical operators. However, it is worth noting that the basic ‘building blocks’ of math and logic have instrumental applications, from which more abstract operations can be extrapolated. When math and logic become too disconnected from the instrumental, as in some discussions around infinity (Wolchover, 2013), controversies of a more philosophical nature tend to arise.
The Objection from Natural Language
Some may argue that since the vocabulary of philosophy consists mainly of ordinary words from natural language (a language that has developed organically through common usage, such as ‘everyday’ English or German), then on the assumption that a natural language is largely semantically determinate, the bulk of philosophical discourse should be too.
The difficulty with this argument is that even on the (very generous) assumption that philosophy can be conducted entirely in the vocabulary of a natural language, the use of expressions from a natural language does not entail the semantic use of a natural language. Even if the tribe in the example had borrowed English words (perhaps from a passing explorer) and used them grammatically, it does not follow that the words are being used in the same sense as they are used outside the tribe, or in any sense at all.
In the absence of instrumental uses, the tribal ‘language’ lacks non-subjective criteria by which the meanings of its expressions can be fixed. Such a discourse does not mean anything determinate, despite using only natural language expressions. On the same grounds, since philosophical discourse is largely non-instrumental, the claim of general semantic consistency between natural languages and the language of philosophy is not demonstrable. Therefore, much like the tribe’s ‘language’, philosophical discourse is, by and large, semantically indeterminate.
The role of context
Some may object that many words in ordinary language (such as ‘game’ or ‘democracy’) are not very determinate in their meanings either, but are usually regarded as meaningful. Why, then, are philosophical expressions singled out as problematic on grounds of semantic indeterminacy?
In most instances of everyday non-philosophical discourse, procedures exist to clarify meanings for (in the sense of ‘in’) a particular occasion of use. Such clarification proceeds either by cursory definition or some form of ostension, and usually succeeds in settling what an expression means for that occasion, in terms of allaying any doubts on that occasion as to how the expression applies (by tacit or explicit reference to the context of use). Through such ad hoc procedures, words like ‘game’ or ‘democracy’ can take on determinate applications in specific settings, even if the general meanings of such words are vague.
For example, someone might say, “This isn’t a game,” when the context clearly indicates that she means, “This is serious business.” In this instance, the otherwise vague term ‘game’ takes on a fairly precise meaning. If her interlocutor fails to catch that meaning, the speaker could follow with a cursory definition such as, “This is serious business.” This may well be sufficient to clarify what she meant on that occasion. Of course, this cursory definition is neither necessary nor sufficient to define all and only games. Much of the explanatory work, of highlighting the relevant sense of ‘game’ out of a nebulous cluster, is done by the context of use.
Attempts at linguistic clarification in philosophy are of a different character. They typically aim to arrive at once-and-for-all general definitions (of expressions such as ‘knowledge’ or ‘truth’), that leave no room for exceptions, and apply uniformly across a wide range of possible settings. Such attempts at clarification tend to generate even more disagreement on what is meant; the opposite effect to clarification in non-philosophical contexts. The radical non-instrumentality of philosophical discourse leaves room for only general meanings; which are indeterminate in the absence of an actual occasion of use.
The question therefore arises, how does philosophical discourse demonstrate semantic consistency with non-philosophical discourse, when the contexts of use for natural language terms are missing in philosophy? How can philosophers ensure that expressions such as ‘mind’, ‘matter’, ‘time’, ‘truth’, ‘equality’, ‘justice’, ‘good’ or ‘meaning’ are used in philosophy the same way that they are used outside philosophy?
Semantic consistency is crucial if philosophers wish to claim that when they are talking about the referent(s) of a certain expression (such as ‘truth’), they are talking about what non-philosophers generally talk about when they use that expression. To avoid confusion, any relevant difference between philosophical and non-philosophical usages of the same term would need to be pointed out. The difficulty is, there is no evidence that most words in a natural language each have a single meaning that is operative in every context of use.
Context seems to play a key role in picking out the particular sense (out of a complex network of connotations attached to a word) that suits a specific occasion, as in the example of ‘game’ above. Furthermore, there is no reason to think that a single definition, however complex, could supplant the role of non-verbal environmental cues (that is, what is observed to be happening other than language) in indicating which sense is relevant in a given scenario.
Those cues are, of course, largely missing in philosophy. Given its purely discursive approach, philosophy is limited to verbal methods of fixing sense, and there is no indication that those methods work for the purpose of ensuring semantic consistency within philosophy, let alone between philosophical and non-philosophical discourse.
For example, one could use an expression such as ‘time’ in the abstract, for instance by saying, “I need more time to finish this task.” But in principle, one could render the abstract concrete by using a clock to show how much time is needed. But if one were to ask, “Is time linear or circular?”, not only are the usual time-related activities (and their related nonverbal cues) not relevant to settling the question, but the further question arises, “Are we still talking about time, or instead a model, picture or description that relates to it, but is not itself time?”
We could refer to time as flowing like a river, or talk about instances of time as ‘slices’, but in doing so, are we talking about the same thing we talk about when using a clock to tell the time and measure time? Or are some of those ways of talking metaphorical, and others literal? Furthermore, are some of those ways unique to philosophy, and therefore possibly not semantically consistent with any non-philosophical usages of ‘time’?
In doing philosophy, we soon discover that there is no straightforward answer to these questions, in relation to most, if not all, expressions of philosophical interest that come from outside philosophy. It follows that much of philosophical discourse is semantically indeterminate, in the sense that the semantic consistency of expressions is often indeterminable, not able to be definitely ascertained. This is because the bulk of criteria from outside philosophy for determining semantic consistency cannot be applied in philosophy, and it is far from clear what other criteria could take their place.
On the other hand, semantic inconsistency is much easier to demonstrate in the abstract (usually by way of ‘thought experiments’) than semantic consistency. Much of philosophy consists of arguments that a model, picture or description of x is inconsistent with, conflicts with, or is misleading about, what ‘x’ refers to outside philosophy.
For example, against the view held by some philosophers that ‘good’ is not descriptive but commendatory, Peter Geach famously argued that “if I call a man a good burglar or a good cut-throat I am certainly not commending him myself” (1956, 36–37), thereby pointing out an inconsistency between a philosophical description of what ‘good’ means and a possible use of ‘good’ outside philosophy. However, showing that ‘x’ is semantically inconsistent in some respect does not amount to showing that it is otherwise semantically consistent.
Having said that, semantic indeterminacy is not always a fault. Some informal, creative or artistic linguistic practices, for example casual conversations, works of poetry, literary fiction, inspirational prose, and even political speeches, tend to be more tolerant of semantic indeterminacy, with a stronger emphasis on pragmatics, style, and affective response, relative to instrumental communication. The same may be said of psychotherapy, in which therapeutic efficacy takes precedence over semantic precision. Hence, nothing in this article precludes reading philosophy as literature, therapy, or even poetry, which despite (or because of) its semantic indeterminacy, offers insights that are subjective and often ineffable, but nevertheless valuable in the reader’s eyes.
Whenever semantic precision matters, instrumental uses of language tend to come to the fore. Philosophy, at least in the analytic tradition, claims to place great emphasis on semantic precision; yet in terms of non-instrumentality and semantic indeterminacy, it has more in common with fiction, poetry, or literary prose than with the scientific genres that many philosophers claim to emulate. Not all philosophers object to this characterization of philosophy; probably fewer do so in the ‘continental’ than in the ‘analytic’ tradition.
Importantly, this critique does not invalidate ‘negative philosophy’, the practice of pointing out obvious inconsistencies, contradictions, unsupported premises, misleading analogies (as in the above-mentioned Geach example), circular reasoning, and other fallacies of an argument. Having said that, it is notoriously difficult to produce a positive thesis out of negative philosophy, which is in essence an exercise in fault-finding. An academic discipline cannot make progress exclusively by shooting itself in the foot. At the least, it has to produce a foot to be shot at.
A classic example of this kind of ‘sniping from the sidelines’ is Gilbert Ryle’s The Concept of Mind (1949), a survey of errors arising from the use of misleading psychological metaphors. However, there is always a risk that if an attempt at negative philosophy is followed by a purely theoretical debate back and forth, it could develop into an extended discursive exchange infected with semantic indeterminacy. To some extent, that fate befell the philosophical literature on The Concept of Mind.
This article is, arguably, an example of negative philosophy, in asking how philosophy differs significantly from the tribe scenario in its (in)ability to show semantic consistency, in contrast to how that consistency is demonstrable outside the discipline (for example by scientists, or even builders constructing a house). In so far as this thesis is negative philosophy, it may suffer the same fate as other texts in that genre, in being buried in philosophical ‘debate’ (although so far, it has been largely ignored). But like The Concept of Mind, this work may yet find a few readers agreeing with the author’s conclusions. When the stakes are high, as they are in this case given philosophy’s prominence in the public square, we cannot afford to sit on the fence for lack of absolute clarity.
The Objection From Memory
Some may argue that even if it cannot be proven that everyday expressions in philosophy mean the same as they do in the larger discourse, we have good reason to think they probably do; because philosophers are hardly likely to forget the proper use of ordinary language while doing philosophy. So, it is argued, the bulk of philosophical discourse, the part in a natural language, is semantically determinate (though technical expressions may contain obscurities, which can perhaps be weeded out).
Such an objection is simply beside the point. Consider the scenario of the tribe earlier in this text. Assuming the tribes-people had borrowed English expressions, and had perfect memories, would it follow that their utterances were semantically determinate? If they were not using the expressions instrumentally, what could they have been remembering? Perhaps grammatical rules and other conventions about how the expressions are organised vis-a-vis other expressions (for example, that you can say “cup of tea” but never “tea of cup”).
However, as the scenario illustrates, even the perfect application of such rules cannot, on its own, produce semantic meaning. Something is clearly lacking in the tribe’s use of English expressions, something necessary to semantic determinacy, but there is no need here to identify that missing element. All that is required, for the purpose of this work, is to recognize that what is lacking in the tribe scenario is also lacking in much of philosophy. In the absence of relevant disanalogies, it follows that philosophical discourse is also semantically indeterminate.
Let us grant that philosophy is conducted entirely in the words and grammar of a natural language, such as German. Even so, it can never be established that philosophers are using German with any more semantic determinacy than the tribe in the above scenario. After all, both the philosophers and the tribes-people are confined to the same criteria for ‘correct’ language-use, namely intuitions about how words can and cannot be organised vis-a-vis other words.
If the tribe’s use of language lacks semantic determinacy, why is philosophical discourse deemed otherwise? How are philosophers remembering to ‘mean the same thing’ when using natural-language expressions in philosophy? Is it not the way the tribes-people are remembering to use English words the ‘right’ way in their non-instrumental quasi-language?
The crux of the problem is not the failure of memory, but the absence of criteria by which to make a selection from memory in a way that constitutes ‘making sense’. Before memory can come into play, the concrete circumstances have to exist in which it makes sense to say that one is remembering to use a word with the same meaning on one occasion as it was used in another. In the case of the tribe, that context is missing, and the criteria they employ are thereby under-determinative of meaning. By the same token, is the context not missing in philosophy? Such a possibility is raised by Stanley Rosen:
We do know what we are talking about (and how to correct errors when we make them) in ordinary language; we cease to be talking about anything in particular when we shift to a purely formal language, and so it becomes literally true that we do not know what we are talking about, unless of course we are talking about the symbols and syntax of the formal language itself. But this, however impressive from a technical stand-point, is not very useful either to the average citizen or to the philosopher. (Rosen 1995, 45)
The ‘language’ of the tribe and by the same token, the ‘language’ of philosophy, is such a formal language: in which there is a disconnect between the syntax of the formal language and the semantic content of the natural language that the formal language claims continuity with.
There are limits to how novel a situation can be before the use of certain words makes no determinate sense, even if they appear to be used ‘correctly’. To indulge in a metaphor, can one remember to use a fishing rod ‘the ordinary way’ in a desert? There is a sense in which even in the desert, one can go through the motions of fishing ‘the ordinary way’ (for example, by holding the rod in the typical fashion), or even do it ‘the wrong way’ (perhaps holding the rod by the narrow end). But it would be less misleading to simply reply, “There is no ordinary way to use a fishing rod in the desert, because you don’t fish in a desert.”
Philosophers could certainly go through the motions of ‘correcting’ one another’s use of ordinary words based on the results of thought experiments, as the desert fishermen could for one another’s use of fishing rods (and the tribes-people for one another’s use of English expressions). But the ability to check one another’s use of everyday language in this way, from memory alone, does nothing to validate the claim that everyday words are used in their natural or ‘ordinary’ sense in philosophy.
Unlike sporadic bouts of purely discursive activity (such as casual conversations), philosophy seminars are not simply ad hoc departures from instrumental language-use. Philosophy is systemically and consistently isolated from instrumental discourse, from the concrete circumstances in time and space that give words determinate meanings. Even if philosophers have perfect memories, and all words in philosophy are taken from natural languages, it would not matter. The memory of how to use an ordinary expression is of little use in an extraordinary setting.
In typical non-philosophical settings, real-world empirical criteria would filter out inappropriate linguistic usages (for example, someone referring to a fork as a ‘spoon’), but in philosophy there are no such tests. There are only logico-grammatical tests in the a priori realm of thought experiments; tests which are under-determinative of linguistic sense because in them, philosophy does not use language to do things in the physical world (other than manipulate purported symbols), or demonstrate sufficient continuity with natural languages (beyond using the same words in the same order) to ride on the latter’s claim to meaning.
The Socio-Psychological Dimension
It is hoped that the preceding remarks have shown that philosophy’s semantic determinacy is very much up for contention. It follows that philosophy is by and large semantically indeterminate (unless one simply takes a ‘leap of faith’ that it is determinate). Philosophers are attempting to pull themselves up by their own bootstraps, using semantically indeterminate discourse to debate the discourse’s semantic determinacy (when they bother to debate it. Most philosophers prefer to skirt the problem, i.e. take the ‘leap of faith’ approach, or assume that the problem is only a local one within certain areas of philosophy, such as whatever they label as ‘metaphysics’). But the problem is not simply that the debate is inconclusive, but that the attempted solution, namely more philosophy, is self-defeating.
Because of its purely discursive methodology, philosophy is especially prone to socio-psychological factors that render it virtually impossible for philosophers to demonstrate that they have arrived at a solution. In a paper entitled ‘Sociology as a Private Language’ (1983), the sociologist Arthur Brittan observed that the specialized technical vocabulary of his discipline functions in some respects as a ‘private language’ whereby, as he put it, “Academics become locked into their own language games, which eventually are externalized and given shape in the texts [which] themselves are then defined as the arbitrators of . . . experience” (1983, 583).
Brittan’s approach was an informal one. In his own words, “I am not concerned with the felicities and ambiguities of linguistic philosophy or conceptual analysis” (1983, 581). Rather, he sought to warn in general terms that by focusing on the exegesis of its own increasingly technical and self-referential literature, sociology risked losing perspective on the social issues it sought to illuminate. This section attempts to frame Brittan’s thesis as a more formal, socio-psychological argument reinforcing the more philosophical case presented thus far.
This formal argument proceeds by contrasting philosophy’s methodology with that of the sciences (not to idealize science, but to use it as a foil to highlight certain structural flaws in philosophy). Scientific and technical disciplines have a ‘building block’ configuration, in which knowledge is assembled out of predictions that have been rendered authoritative by replication, often across disciplines, and in many cases by non-experts, such as amateurs and users of everyday technology.
In general, expressions in such a ‘building block’ discourse have highly determinate meanings, in so far as the results they describe are widely replicable under critical scientific scrutiny (to preempt any charge of ‘scientism’, it is worth mentioning that this claim is purely about the semantic determinacy of building-block discourses, not their veracity). In contrast, philosophy has an ‘all or nothing’ configuration, in which virtually all relevant arguments (with the exception of those already proven scientifically, logically or mathematically) are critically examined from first premises in any new inquiry. The reason for this practice is beyond the scope of this article, which is chiefly concerned with what is entailed by the ‘all or nothing’ approach, within a purely discursive discipline.
It follows that any rigorous philosophical inquiry calls for in-depth knowledge of the relevant literature (knowledge possessed almost exclusively by philosophy postgraduates, preferably PhDs). Some may argue that this is an unremarkable feature of most academic disciplines. However, practitioners in most other disciplines are not expected to have a detailed grasp of every aspect of their inquiry. For example, a biologist does not usually need to replicate all the secondary research that underpins her own work. She can ordinarily trust in cross-replication by others to underwrite the reliability of many findings in her literature survey, some of which may lie outside her area of expertise.
On the other hand, a philosopher would be expected to critically review every work in his bibliography ‘from the ground up’. Because such an effort requires expertise in philosophy and little else beyond common knowledge, its practitioners rely for critical feedback almost exclusively on other philosophers who have been similarly trained, and correspondingly, discount the views of anyone without such training (with the rare exception of opinion polls of what ‘ordinary people’ believe).
This intellectual insularity renders philosophy into a ‘black box’; in that the philosophically unqualified are unable to make an informed judgement (in the eyes of philosophers) as to the quality of its contents, beyond the most peripheral observations. Thanks to the ‘all or nothing’ approach, the conclusions of philosophy are (ostensibly) not testable without in-depth knowledge of the methods by which they are arrived at; knowledge which can apparently only be gained through an extensive program of study, in philosophy. As a result of this epistemic circularity, the influence of cognitive biases and conformity effects within the discipline cannot be independently assessed (more on that later).
Given that philosophy is a ‘black box’ in the above sense, it is impossible to make a judgement that is both informed and disinterested as to whether its discourse makes sense. The rational stance towards philosophy is therefore one of agnosticism as to its semantic determinacy: with the exception of sub-disciplines under ‘philosophy’ that have a building-block configuration (for example, some parts of logic, experimental philosophy, metaphysics or cognitive science).
Some may object that the simpler arguments in philosophy offer windows of scrutiny into the discipline for the educated non-expert. This claim is misleading, because most major arguments in philosophy are accompanied by counter-arguments which are prima facie equally plausible. It is impossible for even diligent non-experts to make an informed judgement as to which philosophical arguments ‘work’, without detailed knowledge of the overarching debate. Even within professional philosophy, there is considerable disagreement on the soundness of most of its substantive arguments.
In the case of ‘building block’ disciplines, no such in-depth knowledge is required for a non-expert to know, for example, that radios generally work as they should, or appendectomies are usually successful. Collectively, such publicly intelligible displays of advanced technical know-how demonstrate that the relevant theories are not semantically indeterminate, or in less technical parlance, ‘full of hot air’. Scientists rarely acknowledge the fact, but the non-expert public is heavily involved in keeping non-sense out of the sciences, effectively acting as a second, informal, layer of quality control after peer review. Unfortunately, not so in philosophy, where one has to demonstrate an expert knowledge of the subject for one’s opinion to count beyond the most basic level.
Cognitive bias and conformity
As mentioned above, philosophy’s intellectual insularity renders it vulnerable to cognitive biases and conformity effects that could distort the discipline’s ability to judge the quality of its output. The idea that philosophers are likely to retain their non-philosophical sensibilities when doing philosophy (and are therefore able to detect deviations from ‘ordinary’ meanings in philosophical discourse) fails to account for the psychological effects of conformity to philosophical modes of communication.
Such effects are exacerbated by the aforementioned non-instrumentality of philosophical discourse; which offers no opportunities for feedback from beyond the ‘ivory tower’ on the quality of philosophical research, through the equivalent of public appraisals of scientific predictive power and technical efficacy. Because the criteria for success in philosophy are (for all practical purposes) internal to its discourse, students have to master the relevant vocabulary and literature in order to be judged competent to critique the discipline’s output, rendering external feedback from non-experts largely irrelevant.
The resulting epistemic circularity strains the ability of philosophers to maintain semantic continuity between ordinary and philosophical discourse. In the course of studying philosophy in an insular and hierarchical academic setting, a scholar’s judgement is subjected to conformity effects. The intensive process of learning new philosophical terminology and usages (as well as ‘unlearning’ the old) induces a loss of perspective, which erodes the value of prior assumptions as a guide in doing philosophy (especially when the discipline aims at questioning those very assumptions).
Even when engaging alone with philosophy through self-study, a person’s judgement is always subject to cognitive biases, distortions in judgement arising from innate human dispositions that may have once conferred an evolutionary advantage (for example, the ‘optimism bias’, a tendency to rate one’s own abilities and the likelihood of future success more highly than warranted; a trait which possibly aided group survival). Psychologists have identified over fifty types of cognitive bias (Benson, 2 September 2016), which can often distort an individual’s perception of whether he or she is making sense (Fig. 1). These biases are reinforced by the ‘bias blind spot’, the overarching bias of believing oneself to be less vulnerable than others to cognitive bias.
A fundamental role of institutionalized academic disciplines is the mitigation of cognitive bias; through collective peer-criticism further grounded in feedback from disinterested outsiders, less handicapped by intellectually compromising associations with the discipline in question. Intellectual accountability to external observers is key to the mitigation of conformity effects within a discipline; such as susceptibility to majority peer influence, and deference to authority figures. Peer review is an empty gesture if the process is effectively opaque and unaccountable to those outside the discipline.
Without meaningful external scrutiny, there is little to prevent the transmission of biases across a discipline through its authority figures. Such scrutiny cannot rely solely on expert peer review, because one’s peers may approve of one’s output for a variety of reasons, personal, political or professional, that have nothing to do with its semantic determinacy. Rather, effective intellectual scrutiny encompasses the collective and cumulative judgments of various disciplines and numerous ordinary people, including the readers of this work. Every time we take a paracetamol, drive a car or draw up a will, we are participating in a mass quality-control exercise that underwrites the semantic consistency of the relevant technical discourses.
To demonstrate the seriousness of the threat posed by conformity effects within a discipline, a few findings from experimental psychology are outlined here. The widely replicated Asch experiments on social conformity have an immediate relevance to how academic philosophy is conducted. In the classic experiment by Solomon Asch, a group of volunteers was asked which one of three lines on a card was the same length as a fourth reference line on another card (Fig. 2). The subjects were to take turns giving their answers. Under normal circumstances, an average of only one person out of thirty-five gave a wrong answer.
However, in some experiments one participant was unaware that before his turn, the others (who were conspiring with the experimenter) would each deliberately call out the same wrong answer. In those ‘rigged’ experiments, about one-third of genuine subjects gave the same wrong answer as the conspirators. In this one-third, the subjects chose to second-guess the evidence of their own eyes and draw a different conclusion, based on what the majority appeared to believe (Asch, 1951). The Asch experiment demonstrates the ‘bandwagon effect’, the tendency to do or believe something because others are apparently doing or believing the same thing.
The setting of the Asch experiment is not altogether unlike a philosophy seminar; where the lines on the cards are replaced by (invisible, and therefore more elusive) abstract concepts, and the role of the experimenter is filled by an authority figure in the form of a senior academic who ‘steers’ the discussion. In a famous experiment (Fig. 3) on obedience to authority, Stanley Milgram (1963) found that when asked by an experimenter, most volunteers were willing to give very severe electric shocks to a stranger (really an actor pretending to be shocked).
Milgram noted that when the volunteers were allowed to choose the voltage, most stopped at the lowest levels. He concluded that the experimenter, an authority figure represented by an academic, was able to override the conscience of most volunteers. In academic philosophy, lackluster participation usually elicits a price, in terms of advancement in a degree program or progress towards tenure. The incentive is great to avoid questioning the basic assumptions of the discipline for the sake of ‘joining in’. Those with fundamental doubts are likely to leave the discipline, depriving it of internal critics; while the views of external critics are generally discounted as lacking in expertise.
The more we sacrifice time and effort on an activity, the greater our desire to defend it. Psychologists call this tendency ‘effort-justification’. In one study on effort-justification by Aronson and Mills (1959), volunteers underwent either a mild or severe initiation ceremony to join the same activity. Volunteers who got in ‘the hard way’ rated the activity more highly. The considerable time and effort required to learn philosophy is an incentive to effort-justification, which fosters a reluctance to question the fundamental assumptions ‘justifying’ the effort.
The conformity effects in the Asch and Milgram studies were severely reduced in the presence of visible dissent. Some may argue that since philosophy thrives on dissent, conformity within the discipline is likely to be negligible. However, there is very little disagreement in philosophy on the general semantic determinacy or relevance of the subject itself. Anyone who doubts the utility of philosophical discussion is likely to leave of her own accord; ensuring that basic methodological assumptions remain largely unchallenged within the discipline.
In technical subjects, there are mechanisms in place to mitigate the effects of both cognitive bias and institutionalized conformity. Enterprises such as the natural sciences, law, history, and even various arts and crafts are intellectually accountable to the lay non-specialist in a thousand small ways for what they say and do. The practitioners in these disciplines have (by virtue of their subject-matter and/or methodology) an institutionally self-imposed obligation to operate according to standards shared across a diverse range of epistemic communities outside the discipline, including the non-expert public.
Certain conditions have to apply for non-specialist feedback to contribute effectively to countering cognitive bias and institutionalized conformity. These conditions include (but are not limited to) the discipline’s self-imposed intellectual accountability to outsiders and non-experts (through their evaluation of the discipline’s predictive efficacy), freedoms of expression and information, a ‘building block’ configuration to research findings (allowing each block to be independently tested through a ‘division of labor’), a rich network of causal or inferential connections between the discipline’s discourse and predicted outcomes, and pressure within disciplines towards greater predictive power and technical effectiveness.
A full analysis of these conditions is beyond the scope of this work, but what is relevant here is that even in sub-optimal conditions (as is usually the case), feedback from non-specialists on predicted outcomes sets limits to what technical discourses can get away with in terms of semantic inconsistency, incoherence or vacuity. The absence of this feedback mechanism in philosophy (essentially due to the discipline’s purely discursive methodology, with the results cited above) leaves it with little defense against cognitive bias and institutionalized conformity.
Most philosophers would probably greet this article with extreme skepticism. They may contend that if philosophy is generally semantically indeterminate, there should have been a point where ‘the penny dropped’ among many of its practitioners. There are several probable reasons why this has not occurred. Most students begin philosophy by reading non-technical introductory texts in plain language, and the default attitude to any grammatical text in a natural language is to assume that it must be saying something determinate.
It is difficult to shake off that conviction, unless one is constantly mindful of the non-verbal environmental conditions that underwrite that sentiment, which are missing in philosophy. Even though substantial cynicism towards philosophy exists among outsiders, most lack the conceptual tools to explain their dissatisfaction to philosophers (who generally ignore the views of non-philosophers anyway, as stemming from ignorance and prejudice). Many students and scholars do abandon the discipline, but their criticisms of it are often ascribed to ignorance or an anti-philosophical bias.
One significant practical implication of philosophy’s semantic indeterminacy is the restricted scope of its contribution to the public sphere. A formal philosophical education can impart valuable general skills of reasoning, argument and textual research, as well as some incidental knowledge (for example, of various logical fallacies), but it does not grant the philosopher a privileged general epistemic status (for instance, as a ‘moral expert’) over non-philosophers. Stanley Fish has written eloquently on this issue, in a statement which also sums up the main point of this article:
Now it could be said (and some philosophers will say it) that the person who deliberates without self-conscious recourse to deep philosophical views is nevertheless relying on or resting in such views even though he is not aware of doing so. To say this is to assert that doing philosophy is an activity that underlies our thinking at every point, and to imply that if we want to think clearly about anything we should either become philosophers or sit at the feet of philosophers. But philosophy is not the name of, or the site of, thought generally; it is a special, insular form of thought and its propositions have weight and value only in the precincts of its game.
Points are awarded in that game to the player who has the best argument going (‘best’ is a disciplinary judgment) for moral relativism or its opposite or some other position considered ‘major’. When it’s not the game of philosophy that is being played, but some other: energy policy, trade policy, debt reduction, military strategy, domestic life, grand philosophical theses like ‘there are no moral absolutes’ or ‘yes there are’ will at best be rhetorical flourishes; they will not be genuine currency or do any decisive work. (Fish 2011)
Philosophers often have insightful things to say, and have made significant intellectual contributions to a wide range of disciplines. But the question before us is whether such insights result from doing philosophy, or simply from the exercise of general intellectual aptitudes. After all, philosophers are people; and some people are wise and insightful. But they are not so by virtue of doing philosophy.
Having said that, as previously mentioned, there are ways (such as the literary, therapeutic or poetic) of reading philosophy that take its semantic indeterminacy into account. Nothing in this article detracts from those approaches. A useful way to characterize such ‘meta-linguistic’ readings is via Martin Heidegger’s famous distinction between two modes of encountering objects (in this case, texts), as ‘ready-to-hand’ and as ‘present-at-hand’ (1962).
When an object (say, a hammer) is viewed as ready-to-hand, our attention is focused less on the object than on what we are doing with it (hammering). When ready-to-hand, the object is actively being used, and the user is not self-consciously attentive to features of the object, but rather to what it is being used for (the task at hand). In other words, the object becomes almost invisible or transparent to the user’s reflective attention. In contrast, when an object is viewed as present-at-hand, the object is the focus of our self-consciously theoretical or reflective scrutiny; not as a thing to be used, but as something that needs to be categorized, organized, repaired or otherwise ‘dealt with’ before it can (if at all) be ready-to-hand.
Similarly, when a text is ready-to-hand, our reflective attention is focused less on the medium (the words, grammar, logic, pragmatics and prosody) and more on the message (what is being conveyed by the text). The medium becomes virtually transparent to self-conscious reflection. This is how we use language most of the time. It seems fair to say that poetry, for example, is not a genre that is meant to be read as ready-to-hand. In the words of the poet Archibald MacLeish, “A poem must not mean, but be” (1985, 106). When we read a poem, our reflective attention is focused primarily on the text itself; and what the poem conveys is usually far from clear (hence our focus on the text).
In poetry, it is the play of words that is center-stage in our mind’s eye, and any ‘point’ beyond that is only indirectly and tentatively apprehended (usually even by the poet). So in its archetypal form, poetry is language present-at-hand. If a poem becomes mundane or ‘dead’ to us (for example, if it is just a shallow limerick, or is over-familiar to us because we had to memorize it), then it may become ready-to-hand, but that is not the reading strategy that is appropriate to the genre. Unlike technical or scientific texts, poetry can never really become transparent or ‘second nature’ without losing something of its value or point.
The same goes for philosophy. It seems a fair observation that philosophical texts are typically the most present-at-hand texts. As many philosophers will concur, they are meant to be read that way; not as putative statements of fact as in science or math journals. As mentioned earlier, some objects are present-at-hand provisionally, in the sense that they need to be categorized, organized, repaired or otherwise ‘dealt with’ for the purpose of being made ready-to-hand.
Many (if not most) philosophers do philosophy with the intention of converting its present-at-handedness into ready-to-handedness (through conceptual analysis or explication). Their ultimate goal is to write philosophical texts that can be read as statements of fact with the same authoritativeness as well-corroborated scientific data. Of course, that goal is never reached, but it informs the ‘scientistic’ way that much of philosophy is done. Because philosophy is purely discursive, and hence generally semantically indeterminate, it is not potentially ready-to-hand. In other words, it is irremediably present-at-hand.
Of course, it is possible to claim that because physics used to be called ‘natural philosophy’ and Aristotle made some (mostly bad) scientific observations, some of philosophy can be ready-to-hand. Indeed, this article draws on some arguably ready-to-hand observations by philosophers, such as Ludwig Wittgenstein, Ryle, and Heidegger, alongside works by sociologists and social psychologists. But the fact that some doctors can do a bit of programming does not invalidate the claim that medicine is characteristically not programming.
Viewing the philosophical text as typically present-at-hand is essential to a standpoint that takes account of the text’s semantic indeterminacy. This standpoint renders the interpretation of philosophy as richly problematic and indeterminate as the interpretation of poetry (especially since there will be constant tension and shifting between ready-to-hand and present-at-hand readings; as we try to make sense of the text while grappling with its semantic indeterminacy). Perhaps that is the point of the exercise.
Some of the content in this work is based on the author’s article: ‘Philosophy as a Private Language.’ in Essays in Philosophy, 13 (2012): 54–73.
Aronson, E., Mills, J. ‘The Effect of Severity of Initiation on Liking for a Group.’ Journal of Abnormal and Social Psychology, 59 (1959): 177–181.
Asch, S. E. ‘Effects of Group Pressure Upon the Modification and Distortion of Judgment.’ In H. Guetzkow (ed.) Groups, Leadership and Men. (Pittsburgh, PA: Carnegie Press, 1951).
Ayer, A. J. Interview with Bryan Magee for the television programme Logical Positivism and its Legacy. (London: British Broadcasting Corporation, 1976).
Ayer, A. J. Language, Truth and Logic. (London: Penguin, 2001).
Ayer, Alfred, and Rush Rhees. ‘Symposium: Can There be a Private Language?’ Proceedings of the Aristotelian Society, Supplementary Volumes 28 (1954): 63–94.
Benson, Buster. ‘Cognitive Bias Cheat Sheet.’ Better Humans, 2 September 2016. http://www.psychologytoday.com/blog/in-practice/201301/50-common-cognitive-distortions (accessed June 20, 2016).
Brittan, Arthur. ‘Sociology as a Private Language.’ Poetics Today, 4 (1983): 581–594.
Cavell, Stanley. The Claim of Reason: Wittgenstein, Skepticism, Morality, and Tragedy. (New York: Oxford University Press, 1999).
Fish, Stanley. ‘Does Philosophy Matter?’ Opinionator-NYTimes.com, 1 August 2011. http://opinionator.blogs.nytimes.com/2011/08/01/does-philosophy-matter/ (accessed November 26, 2011).
Geach, P. T., ‘Good and Evil’, Analysis, 17 (1956): 33–42.
Gibran, Ben. ‘Philosophy as a Private Language.’ Essays in Philosophy, 13 (2012): 54–73.
Graham, Paul. ‘How to do Philosophy.’ 2007. http://paulgraham.com/philosophy.html (accessed April 26, 2014).
Heidegger, Martin. Being and Time, translated by J. Macquarrie and E. Robinson. (Oxford: Basil Blackwell, 1962; first published in 1927).
Jackson, Frank. From Metaphysics to Ethics: A Defence of Conceptual Analysis. (Oxford: Clarendon Press, 1998).
MacKinnon, Edward. ‘Scientific Realism: The New Debates.’ Philosophy of Science 46 (1979): 501–532.
MacLeish, Archibald. Collected Poems 1917 to 1982. (Boston: Mariner Books, 1985).
Milgram, S. ‘Behavioral Study of Obedience.’ Journal of Abnormal and Social Psychology 67 (1963): 371–378.
Papineau, D. ‘The Poverty of Analysis.’ Aristotelian Society Supplementary Volume 83 (2009): 1–30.
Passmore, J. ‘Logical Positivism.’ In P. Edwards (Ed.). The Encyclopedia of Philosophy, Vol. 5, 52–57. (New York: Macmillan, 1967).
Rosen, Stanley. ‘The Metaphysics of Ordinary Experience.’ Harvard Review of Philosophy 5 (1995): 41–57.
Russell, Bertrand. My Philosophical Development. (London: Allen Unwin, 1959).
Ryle, Gilbert. The Concept of Mind. (London: Hutchinson, 1949).
Searle, John. ‘Minds, Brains, and Programs.’ Behavioral and Brain Sciences 3 (1980): 417–424.
Williamson, Timothy. ‘Philosophical ‘Intuitions’ and Scepticism about Judgement.’ Dialectica 58 (2004): 109–153
Wittgenstein, Ludwig. Philosophical Investigations. (New York: Macmillan, 1953).
Wittgenstein, Ludwig. Tractatus Logico-Philosophicus. (London: Routledge Paul, 1961).
Wolchover, Natalie. ‘To Settle Infinity Dispute, a New Law of Logic.’ Quanta Magazine (November 26, 2013) https://www.quantamagazine.org/to-settle-infinity-question-a-new-law-of-mathematics-20131126/ (accessed June 3, 2020)