By what James Joyce would call a vicus of recirculation, I find myself once more in agreement with Ludwig Wittgenstein after an unexpected falling-out.
It was my reading of Wittgenstein’s well-known dictum that ‘the meaning of a word…is its use in the language’ along with his notion of ‘language games’ and ‘forms of life’ that sent me plunging back into the preliterate past on what proved a course of unexpected discovery.
Regarding the dictum, I had always taken the emphasis to fall on ‘use’, so that the contrast was between a word actively employed – dynamic, doing something – as opposed to the word in isolation, in repose, as it were, sleeping between the covers of the dictionary. My reading was that ‘meaning’ was not an inherent property of words but something they had only when they were put to use, like a charge they acquired when they were active – and where that happened, of course, was the ‘language game’ or ‘form of life’.
The difference between those two, as I understand it, is a matter of scale: put simply, the language game is the smaller unit, a particular activity defined within the wider context of the ‘form of life’ – which in turn has a range of applications, from all the activities typical of a group or occupation to the culture of a tribe (such as Amazonian Indians or Cambridge dons). In any case, these were places where language was used, and in which the meaning of Wittgenstein’s dictum could be tested.
For example, the ancient practice of buying a bus ticket from a conductor might be classed as a language game, so that the utterance ‘fourpence please’ (together with the proferring of coin) will initiate an activity which culminates in the conductor issuing a ticket to that value; or in Wittgenstein’s ‘building game’ the utterance ‘slab’ will see the builder’s assistant bring him the appropriate item from the pile. Wittgenstein’s point, as I understood it, was that the meaning of the word derived from its context and was inferred from it – in contrast to the conventional explanation that the utterance is a shortened form of a longer one that is ‘understood’ but not spoken, so that ‘fourpence, please’ really means ‘could I have a fourpenny ticket, please?’ And ‘slab!’ means ‘bring me a slab’: in other words, the grammarian’s urge to supply a complete sentence in order to explain the meaning of the word or phrase was not needed: its use (in the language game) showed what it meant.
What caused me to stumble, when I stepped back into preliterate times, was the realisation that words are an artefact of writing (though as a matter of fact, not all writing systems make word divisions). They do not exist as separate elements in speech, which presents as a rhythmic flow of sound. Furthermore, preliterate speech was embedded in, and inseparable from, human activity: it was always part of a language game or form of life, having nowhere else to go – our preliterate ancestors could not contemplate the meaning of the written word nor hear it spoken in isolation (as on a radio programme, say). As such, I realised, speech would be overshadowed by a whole range of accompaniments that were more immediately ‘speaking’ than it was itself: glances of the eye, facial expressions, gestures, posture, bodily movement (one need only think of the effectiveness of mime in communicating to see this).
At first, I took refuge in the fact that ‘meaning’ could be seen, in a sense, as ‘spread across’ the whole activity (or language game, if you like) – any speech was a contributory part to something that was understood as a whole (so that people would ask ‘what is going on here?’ rather than ‘what did he say?’ in order to find out what was happening). However, I gradually realised that I was bringing my own (literate) understanding to bear in doing that: our ancestors would not have had any use for ‘meaning’ as a term: people might have misunderstood situations or others’ intentions, but not their utterances (chiefly because those did not play as significant a part as they do for us: they would not carry the same burden of meaning, since intention was largely conveyed by other forms of expression, and understanding came from paying attention to the situation, what people were doing, rather than to what was said).
This led me, unexpectedly, into conflict with Wittgenstein. Even though it was his dictum about use and his notion of the language game that had set me on my course, I saw now that meaning was an inherent property of words, and that was a change that came about through the discovery of words, as elements of speech, through writing. Put simply, something written on a tablet can be read and understood in itself, without reference to anything external. ‘The cat sat on the mat’ does not require to be checked against an actual situation to be understood: if I know the words and can read, then I know exactly what it means, regardless of whether there is a cat or a mat for it to sit on. In other words, you can know what something means without knowing whether or not it is true, in the sense of referring to an actual state of affairs. I came to see that Language, as we know it today, is not only an artefact of writing (which gives speech a visible, permanent form that can be analysed) but also has a peculiar character: it is what I would call both an abstract entity and a self-validating structure (or whole).
By ‘abstract entity’ I mean the process by which it is transformed from preliterate speech, inextricably entangled with human activity and overshadowed by the more immediate forms of expression that were its invariable accompaniments, into a separate, self-contained entity: through being written down, and so separated from its original context, speech can for the first time be considered as something in itself; at the same time, its visible form allows it to be analysed, so that what is heard as a rhythmic flow can be seen as a pattern articulated from separate elements or words.
By ‘self-validating structure’ I mean the property noted above, that something written can be understood in itself, without reference to anything external, and this because words can be seen as like the nodes or meeting-points in a complex structure (the Forth Bridge, say) where each part is supported by every other part and each node is the meeting point of forces that fix it in place. In the same way, the meaning of a word is determined, not by its relation to anything external, but rather by its relation to all the other words, its place in the language.
[This, of course, is the basic structuralist model, which considers any given sentence in terms of two different planes: the horizontal plane consists of the sentence itself, and how the words in it relate to one another. In an uninflected language like English, position matters: The cat sat on the hat means something different from The hat sat on the cat; in an inflected language like Latin, it is the word endings that indicate the grammatical relation (puella amat nautam, nautam puella amat and amat nautam puella all mean ‘the girl loves the sailor’ but puellam amat nauta means ‘the sailor loves the girl). The vertical plane consists of all the words that could stand in stead of those in the sentence. It is easier to picture this in terms of a one-armed-bandit or fruit machine: instead of a row of bells or lemons, the sentence The cat sat on the hat is displayed in the window with each word on a separate drum or cylinder. The lever is pulled, the drums spin, and a new sentence is formed: A bird nested under an arch. This has the same grammatical form (article+noun[subject]+verb+preposition*+noun[object]) but means something quite different. Thus, we understand any given sentence not only in terms of its actual content, but also in relation to any grammatically similar sentence; to put it in another, slightly puzzling way, the meaning of a word in any given sentence is defined in part by what it is not, i.e. by all the words that could stand (grammatically) in its stead.]
And it was thinking on this that led me, by a roundabout route, a vicus of recirculation, to reconsider what Wittgenstein actually said, and where the emphasis should fall in the dictum I have quoted. Rather than say ‘the meaning of a word is its use in the language’ perhaps one should emphasise that final phrase, ’the meaning of a word is its use in the language.’
This slight shift significantly alters the perspective: where, before, I had seen ‘use’ as drawing attention to the dynamic character of meaning – that it was only present when words were being used, rather than when they were idle – I now saw the dictum as a repudiation of Wittgenstein’s earlier ‘picture theory’ of meaning, outlined in the Tractatus.
In that, the meaning of a proposition is a matter of correspondence: as the arrangement of elements in a picture must correspond to what it depicts (so that one could, as it were, draw lines from one to the other) so too the arrangement of words in a proposition must correspond to, or picture, some state of affairs in the world. In short, the meaning of a proposition depends on an external relation between it and the world. But if we read Wittgenstein’s dictum with the emphasis on the last phrase, it could be understood as saying that the meaning of a word is a purely internal matter: it is defined by its use in the language, not in relation to anything outside the language. A word does not mean by pointing to something outside itself that validates it; its meaning is determined by it relations to other words, those alongside it in the proposition and (as suggested by the structuralist model described above) those that could stand in its place.
So, having arrived at my notion of Language as a self-validating structure or whole and entered into it gladly as one might a public house after a long stravaig on a hot day, who do I find but Ludwig himself already drinking at the bar, having got there a long way ahead of me.
I’m glad to find myself back in his company.
*I realise that grammarians of a more modern stripe would call ‘nested under’ (and ‘sat on’) phrasal verbs, but I am an old-fashioned sort of fellow.