Edinburgh Research Explorer The affective, the conceptual and the meaning of ‘life’ in the stylistics of Charles Bally

The work in stylistics of Charles Bally (1865 – 1947) attempts to analyse that subset of a language system in which meaning is not purely conceptual or intellectual, but has an affective, emotional dimension. It is not concerned with literary language, but with everyday language used in the service of ‘ life ’ , a word which is central for Bally. This paper adds to the de ﬁ nitive study of Bally ’ s stylistics (Taylor 1981) by bringing in material which came to light after its publication, including Médina ’ s (1985) study of Bally ’ s reliance on the work of Henri Bergson, who recon ﬁ gured the affective-conceptual dyad and whose writings are the source, Médina shows, of Bally ’ s use of ‘ life ’ . This paper also adduces more recently published documents on Bally ’ s intellectual and professional relationship with Ferdinand de Saussure, which ﬁ gures prominently in Taylor ’ s (1981) account, and which can now be reassessed in a new light. (cid:1) 2022 The Author. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). My present


Introduction
More than forty years after its publication, Taylor (1981) remains the definitive study of the stylistics of Charles Bally (1865Bally ( -1947) ) and of the theoretical underpinnings of work in poetics and stylistics which followed in its wake without necessarily relying on Bally.If it has not received the full attention it merits, I dare say the fault is in its star, that he was an underling.Bally has always dwelt in Saussure's shadow.The lack of English translations of Bally has not helped, and indeed is A substantial amount of information has come to light since 1981 concerning Bally and his intellectual relationship to Saussure, a relationship which figures prominently in Taylor's account.Médina (1985) has shown the extent to which Bally drew on the work of Henri Bergson. Amacker (1992, 1994, 2006) has published Bally's correspondence with Saussure and documents relating to Bally's attempt to get a Chair of Stylistics established at the Université de Genève, and Forel has produced a magisterial study of Bally's unpublished manuscripts.The conclusions of Taylor's thorough and astute account of Bally's stylistics are not overthrown by what this newer material has revealed.Unsurprisingly, though, the account is both nuanced and deepened by some of this informationindeed it would be astonishing if that were not the case.The aim of the present paper is to revisit Bally's stylistics in the light of what is now known, and to pursue some questions which this deepened perspective brings into relief.

Bally and Saussure
The first sentence of Taylor (1981) reads: 'A theory of style is a theory of communication'.That is a very stylish opening, in the sense that it grabs the reader's attention by making an equation of two concepts generally treated as opposites.Is style not what is more than communication, conceived as purely functional?Architectural style, clothing style, literary style are what are about more than making a standing structure, covering the body, or getting a message acrosswith the proviso that a stylistically minimal functionalism can itself be a powerful fashion statement.
Yet that opening gambit is quite right when it comes to the sort of 'style' implicit in Bally's stylistics.It is about what gets expressed, and in what way, in a given language; and in order to analyse that one has to consider not just what a speaker or writer intends, but what hearers or readers receive and interpret.Alas, by calling what he did stylistique Bally failed abysmally to communicate what he was trying to do.Seeing it in the title of books he published led prospective readers to expect an analysis of literary style, when Bally was adamant that this was of no interest to him.The word had become somewhat fashionable in the first decade of the twentieth century, with chairs of stylistics established in universities in Germany and Italy, and involving an 'aesthetic' dimension in the literary sense.As a classicist, Bally knew that the etymology of 'aesthetic' linked it to feeling of any sort, not particularly artistic feeling, though this had become the word's principal modern meaning outside science (consider that an 'anaesthetic' does not numb one's musical taste).His stylistics was aimed at a language's aesthetic side in the etymological sense, but was received in the vernacular sense.Insofar as it was a theory of communication, it did not start off well.That misunderstanding would lead to neglect of Bally's work, and may indicate as well why Taylor (1981), with 'stylistics' in its title, has not received the attention it deserves.
In Saussurean terms, Bally's stylistics is not about the parole of an individual speaker or writer, but about the langue, the system shared by all speakers.It analyses what each particular language system, such as French or German, possesses by way of 'affective' elements, the subset of the system which links it not just to concepts but to emotions.Taylor (1981) contrasts 'affective' with 'conceptual' elements, where in Bally's work affectif and affectivité are contrasted with intellectuel and intelligence.In English, 'intellectual' does not correspond to 'intelligence' in the way the two words do in French, or at any rate did in Bally's time, when it was in the process of substantial semantic shift (see Paul 2016: 146-148).When Anglophones talk about intellectuals we generally mean pseuds who pose as intelligent because they are not.Intellectual is virtually a term of abuse, and Taylor has done well to replace it.'Conceptual' is a good choice inasmuch as it centres on the concept, the term Saussure used for one side of the linguistic sign before ultimately settling on signifié for this value-through-difference.
Yet conceptual does not resolve another problem with Bally's intellectuel.Neither term actually stands in a strict opposition to the affective.The affective elements of a language are conceptual-plus-affective, as opposed to ones which are conceptual without being affective.To treat the former as simply affective is in line with Bally's practice: he recognised that 'one and the same sign can produce a variety of effects, but the object of stylistics, according to Bally, is the study of the dominant effect of any particular sign or group of signs' (Taylor 1981: 40).So if a sign is predominantly affective, its conceptual function does not cease, just fades into the background.
'It is important ', Taylor (1981: 20) rightly says, 'that Bally's work be seen in the context of the development of Saussurian thought', and more specifically that it represents 'an attempt to explain aspects of verbal communication other than those accounted for by Saussure's bi-planar model of linguistics' (the two planes being the signifier and signified).Taylor (ibid.)identifies 'the central question asked by Bally's stylistics' as how 'particular features of the expression-plane' are linked to something other than 'conceptual' meaning.The word 'meaning' is an obstacle to answering this question, since, as Ogden & Richards made a banquet of pointing out, the meaning of meaning is itself the ultimate question.
No one was more aware of its diabolical complexity than Saussure, which is why he strived so hard to find ways of thinking about linguistic meaning that would provide at least a foothold out of the muddlenot that he ever persuaded himself that he succeeded.As I write this, I am struggling to express a meaning in my mind, using words each of which have multiple meanings in a dictionary; these dictionary meanings are attempts at analysing the uncountable meanings those words have or could have each time someone says them, where it can never been proven that the meaning in the speaker's or writer's mind matches up with the meanings ascribed to them by those who hear or read them.
So much for the easy part, the relatively straightforward 'conceptual' meaning.All the problems it presents are there as well with affective meaning, which adds a vast aesthetic dimension (in the etymological sense, which was also Bergson's sense) that might be psychological, or physio-psychological, and yet which Bally sets out to investigate within language alone.Whatever it is that those 'features of the expression-plane' link to, Bally does not have much to say about its exact nature.When in 1912 Bally proposed the creation of a chair of stylistics in the University of Geneva, he wrote to the dean: Stylistics is a science which, combining the methods of the psychology of language and general linguistics, studies the affective aspects of natural language.I call affective aspects all the expressions of emotions in language.Natural language is spontaneous language in the service of real life (i.e., almost always spoken language) as opposed to the forms of language which do not have this function or have it only indirectly (e.g., literary language, scientific language, etc.) (Bally to Charles Seitz, 17 June 1912, in Amacker 1992: 60) 2Saussure was asked to comment, and raised a concern that 'stylistics' would be understood as something other than what Bally actually did in his analyses.Saussure added that he did not see a real distinction between stylistics and the general linguistics which was part of his own chair.Bally then suggested changing the chair's title to Psychology of Language, and to this Saussure raised no objection.3

Bally and Bergson
To understand the affective in Bally, we need to look to those who shaped his view concerning it, chief amongst whom was, not Saussure, but Bergson.Bally was indeed doing a sort of psychology of language, of a very particular sort: a psychoanalysis of the language itself.Not in a Freudian vein, of course, but a Bergsonian one conducted upon a particular language system fundamentally as Saussure conceived such systems, and using a method which relied largely on comparing one system with another.Here for example is what Bally has to say about deferential second-person pronouns, such as the French use of vous, originally the second-person plural, instead of tu to a single person: The Romans and the Greeks called everyone tu, even their magistrates, their emperors and their gods.We have learned to distinguish people socially by tu and vous, and the details of this distinction are complex: a son calls his father vous, who addresses him as tu, and says vous to his wife in front of others but tu when they are alone.Under the ancien régime, in a time when the hierarchy of classes was very accentuated, German had four forms of address: for 'Do you understand?' one said not only Versteht du?And Verstehen Sie? but Versteht Ihr? and Versteht Er?Only the first two have survived, the third is disappearing, and the last is dead.English, always practical, calls everyone vous (you) and only says tu (thou) to God. (Bally 1926(Bally [1913]]: 84) 4Anglophone learners of French or German soon become aware of the affective force of using the informal pronoun where the formal one is expected, or vice-versa: it can be like a slap in the face, unless the speaker's foreignness is so obvious as to provide a Get Out of Jail Free card.Bally does not analyse what might be at work internally in each of the five language systems considered, as other types of 'psychology of language' would do, apart from the last remark about English being 'always practical', which may or may not be serious.Particularly interesting is how Bally structures his sentences: The Romans and the Greeks called everyone .We have learned to distinguish .A son calls .German had four forms of address .English, always practical, calls everyone .These appear to be equivalent choices for him stylistically, whether in his sense or the usual one.English 'calls', just as a son 'calls', or the Romans and the Greeks 'called'.
This comparative dimension comes to the fore in Bally (1912), a reply to a harsh critique of his books of 1905 and 1909 by a German scholar, Strohmeyer (1910), in which it is clear that Bally has been piqued: his affectivity comes through despite his attempts to temper it.When in 1926 he incorporated this paper into the second edition of Bally (1913), he rewrote it so as no longer to be a polemic with Strohmeyer, who, Bally acknowledges, apparently learned his lesson by following Bally's corrections in the second edition of his own book.But Bally remains focussed on how French and German differ in their affectivity, and this comparison will continue to occupy a key place in his work, including Bally (1932), the best candidate for being his magnum opus, apart from Saussure (1916). 5 Even when he puts forward a 'theory of enunciation' in Bally (1932), it is about how the French or German language enunciates an utterance, not about the speaker, as in the later version of enunciation that emerges in the work of Jakobson, Lacan, Dubois and Benveniste.By no means does Bally deny that affectivity in linguistic utterances has multiple sources, or that context plays a significant role in the recognition of an element as affective or conceptual.Even an inherently affective sign such as paperasserie 'red tape' (on which see Bally 1926: 175-176), with frustration built into it, can lose its emotional charge when a phonologist uses it as an example of a word in which the mute e is dropped thrice.Paul (2016) homes in on what intelligence signified in the last quarter of the nineteenth century.It defined the breach between Hippolyte Taine, whose De l'intelligence (1870) was the vehicle for the diffusion in French thought of associationism (see next section), and of Bergson's subsequent reactions to it.His insightful remark that, for Bergson, 'Intelligence is both opposed and complementary to intuition' (Paul 2016: 153), applies as well to Bally.And that is the nub of the complex relationship of Bally's intuition-based stylistics to Saussure's linguistics, which Bally regarded as the supreme analysis of language from the perspective of intelligence.Both opposed and complementary, which intelligence says is impossible, but intuition says is quite rational and realistic.At least it says so to some of us.
Pace Aarsleff (1978), Saussure did not follow Taine.Despite certain parallels in their vocabulary, the gap between them is wide, and Saussure never cited Taine nor left any record of having read him, whereas Bally cites Bergson directly, and the gap separating them is narrow.But there are indeed ideas shared by Taine and Saussure, and Bergson's strong and sustained opposition to Taine helps to clarify what kept Bally's stylistics from being either identical to Saussure's linguistics or as distanced from it as it might potentially have become.
Bergson used Taine to position himself in opposition to materiality in its usual sense.Intuition, élan vital, life, these concepts have a spiritual dimension.When in the epigram to this paper the translators added 'Our present is the materiality of our life' (see fn. 1), it was to underscore how Bergson denies the usual understanding of materiality as what endures, versus consciousness, which is momentary.For Bergson, time itself is a fact of consciousness, and materiality is within them, not outside them.Consciousness is an aspect of l'esprit 'the mind', hence spirituel, a complex matter since the other French word for the mind, l'âme, also means 'soul', giving it a closer semantic link to English 'spiritual' (see Cassin ed. 2014Cassin ed. : 1009Cassin ed. -1025)). 6 Bergson was thinking in a vein which was not limited to philosophy, but was reshaping physics, in ways that today are generally associated with Einstein and later with quantum theory.But Bergson represented many things to many people.For a reader like Bally, the appeal was that of a modern philosophical-scientific voice advocating intuition over intelligence, spiritual force over mere mechanics.It threw a spanner into any equation of science with (vulgar) materialism.Bergsonianism managed to be both cutting-edge-modern intellectually whilst, beneath the surface, breathing new life into notions which Taine's generation had discarded as unscientific because not materialist.
Bally was caught in a bind.It was not that he was afraid to disagree with Saussure.The problem was that he did agree, when it came to trying to read the mentality of a people from the structure of their language.As he would say in his inaugural lecture on taking up Saussure's chair in 1913, The product of the psychic activity of a social group, a language can, at the very most, furnish some indirect indices about the mentality of a people, but it could not teach us anything certain about its material civilisation.Nearly all representations of reality get subjectively coloured and deformed in penetrating into a language: then the language itself ceaselessly transforms, and what might we expect to remain of the reality in an image which has passed through this double prism?I therefore share my master's scepticism on this point [ .](Bally 1977: 155) 7   In his proposal for the Chair of Stylistics cited earlier, Bally emphasised that it is about language in the service of life, and in 1913 he published a book entitled Le langage et la vie (Language and life).
What in the quote just above corresponds to 'life'?Bally contrasts the mentality of a people with its material civilisation, in the 'vulgar' materialist sense rather than the Bergsonian relocating of materiality within mentality.Bally then talks about reality in a way that is ambiguous: does reality relate just to material civilisation, or to mentality as well?Either way, they would both surely seem to be part of life; and he is saying here that the most we can hope to find in a language are indirect indices and subjectively coloured and deformed representations. 5Bally composed the text of Saussure's Course, working from a compilation prepared by Albert Sechehaye of notes by Saussure and lecture notes taken by some of his students.
6 When in his opening address to the 2nd International Congress of Linguists in Geneva in 1931, Bally (1933: 30)  But in that case, stylistics would be a forlorn undertaking.The conclusion we can deduce from what Bally says here is that neither mentality nor material civilisation is part of life.That seems unlikely, until we consider, first, that material civilisation as Bally intends it here may both frame life and be its product, but still remain distinct from it; and secondly, that mentality is limited to the intellect, the conceptual.It excludes the affective, which however connects to how 'representations of reality get subjectively coloured'.Bally is talking about these things in a very specific way that is easy to miss.

The 'mind-body problem'
Although the distinction of the affective and the conceptual resists pinning down, it is part of a current running through the entire history of linguistic thought, and which is often called the mind-body problem (see Joseph 2018).It sometimes takes the form of a debate, and other times is implicit in assumptions made, questions raised and answers offered.Its centrality in Bergson's thought is evident in the epigram to this paper.In modern linguistics, a version of the mind-body problem underlies the division between phonetics as a study of articulatory and acoustic properties, and phonology as a study of sound systems as they function in the minds of speakers and hearers.It is also reflected in considerations of how minds relate to brains as bodily organs.
The term 'affective' has an ancient heritage.Latin affectio translated Greek path e, what a person undergoes, including feelings in reaction to something experienced.For Aristotle in On the Soul, all speech is preceded by path e, so all encompassing is the conception at that historical stage.Over later centuries, path e and affectio would become more specialised, linked to emotion in opposition to reason; and the English borrowing 'affection', in its everyday usage, is further specialised to warm, positive emotion, as too is 'passion' (from the Greek path e via Latin), sometimes used interchangeably with affection, sometimes distinctively.'Affective', however, has remained a scientific term encompassing emotional responses both positive and negative, as indeed did 'affection' in technical usage at least into the nineteenth century.
In earlier times, when the affections and passions were linked to the bodily humours and the blood circulatory system, their physical nature was more obvious than when the functions attributed to the humours were redistributed, largely to the nervous system. 8The connection of affection and passion to the heart are a historical memory of this.Aristotle (unlike Plato) could not believe that the brain, a cool organ containing little blood and no discernible movement, was the seat of motor functions which had to be transmitted to the muscles, presumably through the circulatory system.Hence his contention that the heart, warm and throbbing, was the body's control centre.Galen of Pergamon would finally overturn that belief, though it still took until the modern period for the electro-chemical operation of the nervous system to be understood.As it was being unravelled, a new theory of mind and body was launched, by John Locke in his Essay concerning Human Understanding.
Associationism, as it would come to be known, developed from Locke's 'association of ideas', itself anticipated by Hobbes and indeed by Aristotle.Locke says that 'Some of our ideas have a natural correspondence and connexion one with another', but his concern is actually with another connexion of ideas wholly owing to chance or custom; ideas that in themselves are not at all of kin, come to be so united in some men's minds, that it is very hard to separate them; they always keep in company, and the one no sooner at any time comes into the understanding, but its associate appears with it.(Locke 1975(Locke [1700]]: 395) The key phrase here is 'some men's minds': that is, not everyone's.The association of ideas Locke is describing is a symptom of a 'disorder in the mind': his one linguistic example is of someone 'surfeiting with honey', who 'no sooner hears the name of it, but his fancy immediately carries sickness and qualms to his stomach, and he cannot bear the very idea of it'an association neither natural nor reasonable.
With Hartley in the mid-eighteenth century, association shifts from being considered a disorder affecting a few to become the cardinal principle of psychology, the source of all passions, affections and dispositions.In his Enquiry into Origin of the Human Appetites and Affections, after considering various contemporary views about how the nerves and brain are constituted, Hartley draws readers' attention to how perceptions of external objects trigger motions from the sensory organs through the nerves to the brain that 'leave there certain faint traces of themselves', and 'in like manner [.] sensations also leave behind them ideas in the mind' (Hartley 1837 [1747]: 67).He describes the association of ideas in terms similar to Locke's except without the negative connotations.'By association I mean that power or faculty by which the joint appearance of two or more ideas frequently in the mind, is for the most part changed into a lasting and sometimes into an inseparable union' (ibid.: 68).He is careful to assign it a specifically physical cause: 'probably, the reason of ideas when once united keeping ever after in company together, is owing to a succession of motions in the body, or, rather, to those motions of the nerves always producing one another' (ibid.).The example he gives is a cross-sensory one: For this is fact; a child has the idea of the sound nurse often presented to the ear, at the same time with the visible appearance of the nurse herself in the eye, and by this frequent conjunction it comes to pass, that the visible appearance of the nurse shall itself excite a faint image of the sound nurse in the ear; and the sound nurse in like manner shall excite a faint image of the visible appearance of the nurse in the eye.And all this seems to be effected by the mutual influence which the motions in the optic and auditory nerves, constituting seeing and hearing, have upon one another according to the laws of matter and motion.(ibid.: 68-69) Hartley's 'faint image of the sound nurse in the ear' is akin to what in the middle ages was called the verbum cordis, and in more recent times 'inner speech'.He identifies it as a physical manifestation, yet, crucially, it is not the sound, but the idea of the sound.Since images of sound and sight are learned by experience, they are not the same from one person to another.Our experience varies and children 'cannot have any Idea, but with some Particularities in the Non-essentials' (1749: 272).
In the second half of the nineteenth century, Alexander Bain put forward an up-to-date version of associationism that took account of contemporary medical science.Bain laid out a series of 'general laws of alliance of body and mind', of which a central one was the Principle of Relativity, 'the necessity of change in order to our being conscious', which he called 'the groundwork of Thought, Intellect, or Knowledge, as well as of Feeling.We know heat only in the transition from cold, and vice versâ; up and down, long and short, red and not redall are so many transitions, or changes of impression; and without transition we have no knowledge' (Bain 1875: 81).He makes the strong claim that knowledge, in effect, equates with difference.
Our knowledge begins, as it were, with difference; we do not know any one thing of itself, but only the difference between it and another thing; the present sensation of heat is, in fact, a difference from the preceding cold.(ibid.) The idea that 'all consciousness is of difference' had been been stated by Mill (1867), crediting it to Hamilton (1859-60).Spencer contains a further elaboration of the steps leading from simple difference to 'consciousness' in the full sense (Spencer 1855: 324-327).Bain adds, as his predecessors did not do, that this principle applies to the sounds of language, and not just to concepts or ideas: 'Our discrimination of articulate sounds is co-extensive with the combined alphabets of all the languages known to us' (Bain 1875: 84).
Bain's associationism became the new modernism of the younger generation of French psychologists, thanks mainly to its being championed by Taine.However, Taine took from Bain just those bits which supported his own 'empiricist' conception of intelligencethe conception in opposition to which Bergson would later construct his own theory of 'intuition'.The distribution of knowledge through the neuromuscular system which is so central to Bain's work becomes inessential in Taine's.In some respects, Bergson, despite his avowed opposition to an associationism which he rejected as excessively mechanical, is closer to Bain than Taine is.
In Bain's associationism, any grounds for a scientific distinction between affective and conceptual elements of a language system are removed, or at least rendered superficial.All knowledge, including that of language, is sensory-neuro-muscular in nature, and begins with the perception of difference, which is the source of consciousness.When Taylor (1981: 40, cited above) says that 'the object of stylistics, according to Bally, is the study of the dominant effect of any particular sign or group of signs', viewed from an associationist perspective, identifying that effect looks like being itself an 'affective' rather than an objective move.
In the part of Le langage et la vie which treats the predominance of the involuntary in speaking, Bally writes: Intelligence in the service of life envelops and surpasses our logic with its geometrical forms: language shows, better than anything else, what that means.One thinks involuntarily of Bergsonian intuition, and language, in its relationship with life, seems to prove Bergson right when he says that 'life overflows intelligence in all regards' [ .]. (Bally 1913: 38-39) 9Bally does not however propose an alternative vision of the language system which would itself dispense with 'geometrical forms' in whole or in part, perhaps turning the affective part into a process, as Bergson might have done.Bally stays close to Saussure in conceiving of the system as being itself inherently processual, generating values on the associative and syntagmatic axesvalues which in some cases, Bally maintains, overflow these axes and connect to life.
Médina shows convincingly that la vie 'life' in Le langage et la vie is a 'totally Bergsonian reference'. 10Bergson's most farreaching cultural contribution is his concept of élan vital, where vital is the adjectival form of la vie.For Bally, life is what is felt, and, obviously, lived.It is synonymous with 'affective', that which 'overflows intelligence' understood as a cold, dry, cerebral calculation of difference.The Bergsonian concept will have an indirect descendence in Merleau-Ponty's corps propre, usually translated as 'lived body', which he substitutes for the traditional Cartesian thinking subject, then replaces in turn with la chair 'flesh'.Bally wants to find in language both the intelligence of a subject and the lived body that responds independently of that intelligence, the pairing which Bergson envisioned, and which Merleau-Ponty, steeped in the phenomenology of Husserl as well as the psychology of perception and the linguistics of Saussure, would later strive to unify.
Herein lies an explanation of much of the gap between Bally and Saussure, who was sympathetic enough to the associationist outlook to have applied the term 'associative' to one of the two axes of the language system as he conceived it (the one which, under the influence of the Copenhagen School, would later become better known as the paradigmatic axis).To the parallels detected by Aarsleff between Saussure and Taine can be added parallels with associationists whom Taine did not follow.I have already noted Bain's equation of knowledge with difference, and Saussure's signifier is close to Hartley's 'idea of the sound', except that the latter must differ from individual to individual because the experiences through which they acquired it were unique, whereas the signifier is just that part which is socially shared.Saussure's concept is addressing the implicit question of what we mean when we talk about 'a language', such as French or English or Chinese, which did not fall within Hartley's concerns nor those of later associationist psychology.
Saussure may have first come across Bain in Egger (1881), which impressed him enough to make detailed notes at least about its opening section (see Joseph 2012: 288-291).Egger contends that thought takes the form of 'inner speech', and he is disdainful of Bain for creating what he calls 'a school of touch, or more precisely, a school of muscle' (Egger 1881: 59), associating Taine with it as well.11For Egger, Inner speech is a simple image, a purely sonorous image; in the same way, the outer speech of another heard by us is a simple sensation, purely sonorous; but it is otherwise with our own speech, perceived by our ear at the same time as it is produced by our vocal organs; this time it is a double sensation, simultaneously sonorous and tactile, or, more precisely, a couple of sensations.(Egger 1881: 75-76) 12Bain did write about a silent interior 'nervous' speech (1855: 334), and called the mental recollection of language 'a suppressed articulation, ready to burst into speech' (1875: 91).But for Egger this amounted only to a 'brief mention'.
Certain aspects of Egger's book seem to prefigure Saussure's conception of the linguistic sign, including this 'sonorous image', which harks back to Hartley's 'idea of the sound'.But Saussure leaves off taking notes on Egger shortly after those notes get to be sharply critical of the theory Egger is putting forward.Egger's subtitle identifies his book as an essay in descriptive psychology, and Saussure will always grant that psychology has a valid claim to authority on the mental functioning of language, different from his own expertise as a grammarian, which is how he usually identified himself.But Saussure's personal decision was to leave psychology to psychologists, and not to pronounce on matters beyond languages themselves.This is why the signified in his model of the linguistic sign involves no claim about real-world referents: not because he thought they had nothing to do with one anotherof course they dobut the signified-referent relationship is for psychologists or philosophers to grapple with.A grammarian can speak with authority only about the signified-signifier relationship.
Bally was not similarly willing to forego psychology.He was determined to bring it directly into grammatical analysis, with a focus specifically on that which, following Bergson, exceeds 'intelligence'.Bally does not take up the bodily dimensions of Bergsonian thought, beyond that central concern with 'life'; he does not consider how the affective might be approached in associationist terms, which appear outmoded in Bergson's perspective because, although they aim to reconcile physicality with consciousness, it is by reducing the conscious to the physical, within a mistaken view of how each relates to time.That left Bally stranded with an affectivity that cannot be divorced from its ancient physicalist heritage, contrasting it with a conceptualism with its own anti-physicalist baggage, on the railway platform of a 'life' constructed to transcend the dichotomy of materiality and consciousness.Whether he missed the train, or it crashed, or his scarf got caught and dragged him under the wheels, he was never going to reach his destination.

Understanding understanding
For Taylor, the problem Bally was trying to solve was how people understand one another.For Bally, this is a different process in French and in German, and by implication in every language; and he aimed to discover what resources the language, as a socially shared system, offers to its speakers and hearers for understanding one another, not just in intellectual/ conceptual terms, but in affective ones.To ignore the latter is tantamount to pretending that life does not exist.This entailed putting aside, first, how these differences may have come about, including any consideration of supposed national psychological traits; and secondly, the dimension of parole, which, although clearly essential, would have to wait for an adequate analysis of the langue without which it could not transpire.Putting aside something essential, even if for valid methodological reasons, constitutes a pretence.
What the development of structural linguistics would reveal is that this pretence had a wide appeal.It is that sort of idealisation which allows for the formulation of nomothetic laws, in an age when scientific glory depended on such laws, rather than on ideographic observations.This despite the fact that observations are immediately scientific, in lending themselves to immediate falsifiability through attempted replication, whereas laws are theoretical, and seeming contradictions can be dealt with by refining them.The problem with 'life' is that it is too messy.Bally tries hard not to make it so, indeed to contain it within the language system rather than dealing with the affective reactions and stylistic differences of individuals.But the inherent weakness in his distinction between the affective and the conceptual is that formulating the conceptual is difficult but imaginable, whereas formulating the affective is utopian to the point of absurdity.Psychology has seen many attempts to formulate the affective, but ultimately their result has been, not universally accepted laws, but at best a renewed vocabulary in which to conduct arguments and debates.
I said in Sec. 2 that a theory of communication, the problem staked out at the start of Taylor (1981), would become the theme of his own work, and indicated that it has sub-themes.'Understanding' is one of these, developed most fully in Taylor (1992).Across the history of linguistic thought can be encountered a trope so culturally engrained as virtually to define what counts as linguistic thought: that for two or more people to understand one another entails a shared language, and conversely, sharing a language entails understanding.Before the modern period, the occasional moments when this is questioned are footnoted, or classified as something other than linguistic thought.These include denials that anything meaningful is being said unless the language is not merely shared, but perfect.For Cratylus, in Plato's dialogue which bears his name, unless a word perfectly represents what it denotes, it is just noise; and according to Aristotle, Cratylus himself finally despaired of words, and would only point at things. 13n the same period, Confucius was similarly repelled by the use of 'drinking horn' for a horn-shaped drinking vessel made, not from an animal's horn, but from brass (coincidentally the metal St Paul chose for the utterances of a speaker without charity). 14But in the history of linguistic thought, St Paul is relegated to religion, and Confucius to philosophy.Cratylus is a curiosity, a literary foil for the exposition of Socrates's belief that we can never know whether a language represents reality perfectly or poorly.Most contemporary linguists do not want to hear about history except as a Whiggish account of steady progress from The-Darkness-That-Was to The-Way-We-Think-Now, and in that perspective, Cratylus, Confucius and St Paul all represent what had to be got over.
Linguistic thought starts, in the Mediterranean world, with grammarthe analogists of Alexandria, Apollonius Dyscolus, Herodian, Varro, Donatus and Priscianand rhetoric; in Asia, with phonology, and the towering work that is the Ast adhy ay ı (Eight chapters) of P anịni, who is not unlike Cratylus or Confucius in his ultimate concern with the correctness of language being the key to true understanding.But unlike them P anịni does not lament the poverty of representation, instead getting down into the fine details, as will Priscian.
It is not as though misunderstanding went unnoticed.Clytemnestra is one of several figures in Greek literature who speak in riddles.St Augustine laments his language's insufficiency to his heart (see next section).Religious disputes over the meaning of particular words and scriptural passages saw heretics burnt and wars waged.'Will no one rid me of this turbulent priest?', Henry II bemoaned or commanded.A later Henry would take England out of Roman Catholicism, in which there was great worry about ordinary people's inability to understand scriptural teaching properly, and into Anglicanism, a non-Roman Catholicism which did not quite share the faith of more strongly protestant sects, including the Calvinism of Scotland, Geneva and other parts of the continent, that God will direct everyone toward right understanding of the divine word.
The names from the past who figure in Taylor (1992) would themselves constitute a course in the history of linguistics, from Plato and Cratylus to Hobbes, the Port-Royal grammarians, Condillac, Rousseau, Herder, Monboddo, Humboldt, Mill, and with Frege and Saussure taking us into the early 20th century.At the centre of the historical panorama is Locke, whom Taylor identifies as the principal figure responsible for putting misunderstanding top of the philosophical agenda, in, funnily enough, the Essay concerning Human Understanding.Locke fits into that Anglican pattern of not assuming that God will sort out understanding, whether for each individual in the Calvinist view, or via top-down authority in the Roman Catholic one.Like other sensualists of his time, Locke trusts the tools we have been given for attaining knowledgeour bodily sensesto let individuals communicate properly so long as they apply principles inherent to the sensory apparatus but usually ignored in language use.'Locke's puzzle' is one of Taylor's contributions to the linguistic-philosophical lexicon.It refers precisely to this gap between the potential and the actuality of language: 'although language serves us well enough for our ordinary purposes, we do not really understand each other' (Taylor 1992: 45, italics in original).Taylor also uses the term 'linguistic scepticism' for this puzzle, although it fits less well.A true sceptic would question whether understanding is ever possible, whereas Taylor's Locke is laying out a practical, if ambitious, programme for resolving the puzzle, not the utopian pipe dream of a man resigned to scepticism.
I have tried to show some of the links which Bally's stylistics has to linguistic thought from ancient times through to the modern period, including to Locke, and the associationism which Locke inspired; and particularly to Bally's contemporary Bergson, whose opposition to Taine illuminates aspects of what separated Bally from Saussure.Taylor (1981) has shown how Bally wanted a linguistics that could penetrate beneath an understanding of a superficial nature, which might have little or nothing to do with the true understanding taking place.It was the same impulse that motivated, in the next generation, Wittgenstein in philosophy, Malinowski in anthropology, Bühler in psychology, and in certain respects, Merleau-Ponty.In the generation after that the names become too numerous to list.At the same time, Bally was devoted enough to the linguistics of Saussure to ensure its transmission; and, seen in the context of the times in which he was conducting his career, his own work pushed the envelope about as far as it could without risking arrest by the postal authorities.

Conclusion
In his inaugural lecture of 1913, Bally (1977: 157) would call Saussure an intellectualiste convaincu, a confirmed intellectualist, meaning that, if teams were being chosen for a tug of war between Taine's materialist intellectualism and Bergson's intuition, Saussure was on Team Taine.Even if, like Bally, he felt the affectivity in language, he needed to explain it conceptually.If fools rush in where angels fear to tread, Bally played the archfool to the archangel Ferdinandfortunately for us, since otherwise he would be an angel long forgotten.Maybe though Bally would have done well to be more angelic with his stylistics, and heed Saussure's cautions.
Consider the title of his 1913 book, Le langage et la vie.Stylistics takes language, the first term, to contain a subset of affective elements which express life, the second term.Is le langage itself a conceptual term, by his analysis, and la vie an affective one?Is vie two distinct signs, a conceptual one when used clinically, and an affective one when used à la Bergson, where it has no meaning that can be pinned down but invokes the universe of feeling and sensation?These are the sorts of questions that Saussure would have seen as too paradoxical and problematic not to sort out before going into print.In Bally's notes for a course he gave in 1927, the year after the second edition of Le langage et la vie appeared, we find what may be a partial reconciliation with his 'intellectualist' mentor: We then understand the deeply intellectual character of all signalisation, and this character is found naturally in the signs of articulated language.We must never forget this, notably when we study expressive language, and the great problem of stylistics is to know how signs of an intellectual nature can move us. 15  This is part of a complex discussion contrasting signs with 'indexes', and it is not a renunciation of the affective-intellectual dichotomy, but a narrowing of it.
Bally's 'life' may have Bergson as its immediate source, but there are more distant echoes.When St Paul writes that 'Though I speak with the tongues of men and of angels, and have not charity, I am become as sounding brass, or a tinkling cymbal' (1 Cor.13:1), does this not imply that the affective dimension, the charity that is in the heart, is not some trivial addon to language and communication, but essential to it?It can be read as a stylistically spectacular version of St Augustine's lament 350 years later 'that my language is insufficient to my heart.For all that I comprehend, I want my hearer to comprehend; and I feel myself not speaking so as to bring this about' (Augustine 1915: 18-19). 16Admittedly I tilt the table in translating cordi meo literally as 'to my heart' rather than to the more modern-friendly 'to my mind'.Yet Paul here foresees the modern world, since, even if an ideally representative language existed, interpretative charity would still be required for any hope of people understanding one another.His faith in the possibility of such charity is strong, whereas experience has weakened Augustine's.
Bally's stylistics looks into the heart, not of the speaker, but of the language, conceived not as parole or even inner speech but as represented in a grammar and a dictionary.He grounds his method in the authority of Saussure, even whilst admitting to their shared scepticism about the representation's inherent distortions.Saussure could not see a clear division between Bally's stylistics and the rest of linguistics, unless stylistics was hived off as a branch of psychology; and as for Bally's other main inspiration, Bergson, there is no indication that he would have taken stylistics seriously.Nothing in his writings accords the place to language that Bally gives it. 17If such an enquiry into the heart of the language fails, what can one say except: tu parles, Charles.Or perhaps, c'est 'la vie', mon ami.