Descartes, when laying the foundations of modern Western philosophy, made a praiseworthy attempt to base his “system” on what seemed to him to be the most “indubitable”. But having unfortunately identified this with what he thought he conceived “most clearly and distinctly”, he proceeded to express it by means of his famous “cogito, ergo sum”: “I think, therefore I exist.”
Now we probably all accept that there is an indubitable core to experience - the “something” with which we feel ourselves to be “directly acquainted” - but this is neither clear nor distinct. Nor, for that matter, was Descartes'; none of his interpreters has been able to attach a “clear and distinct” meaning to his “think”, “therefore” or “exist”, while the “substantial” nature of the “I” implied by the repeated use of that word or, in the Latin version, of the first person singular was effectively demolished by Hume's criticism. Furthermore, his formula accepts the validity of the subject-predicate structure of ordinary language and the Aristotelian type of logic which schematises our use of it. His “therefore” conceals the hidden major premise “Nothing can be said to think - or, a fortiori, to do anything else - unless it can be said to exist”. Such “majors” are nothing more than inductive generalisations of prohibitive rules that are commonly accepted by language-users; they are very far from being “indubitable”.
I suggest that what Descartes was trying to express can be far better formulated by the one word “sentence”:- “Experiencing!” This begs no question, and is sufficiently vague to be universally applicable to anything we may wish to express. My “experience”, although ordinary language encourages me to represent it by a noun, seems to me to be an activity, corresponding rather to a verb; its “stuff” does not seem to be any sort of “invariance” but on the contrary to be more fluid than anything I can possibly name.
Now I wish to express a philosophical view which is radically empiricist, that is to say, one which roots all our “knowledge” in “direct” experience. For this purpose I therefore have to devise a form of language which treats change as the “norm”, and invariances as its “accidents”. This is not as difficult as it sounds; all that is needed is to adopt the verbal gerund “experiencing”1 as a “universal subject” which, once accepted, can be taken for granted and usually omitted. Since this is a verb, all its predicates must be adverbial in function, expressing “modes of experiencing”. Our usual vocabulary contains nothing like enough adverbs for the purpose, but by making free use of the suffix “-ly”, which confers adverbial function on whatever precedes it, we can affix it even to substantives, to make it clear that what we are attempting to express is not “them” but, rather, “experiencing themly”.
______________________
1. My views are closely similar to those of William James, but he used the noun “experience”. In my view “experiencing” must be prior to “experience”: “occurring” must be prior to “existence”. Editor's note: The author bypasses the well-know Aristotelian distinction between priority in the order of knowing and priority in the order of being, illustrated by our concluding to the fact of the Moon's sphericity from our observations of its phases - phenomena which follow from rather than cause that satellite's shape.
I will now demonstrate the potentialities of this technique by using it to analyse a “basic statement” of the kind that scientific philosophers regard as “empirical”. These are nearly all attempts to express visual observations, since visual structures tend to be “quasi-permanent” and are the most detailed and complex that are generally believed to be concurrently noticeable by several observers. The one that I choose is a modification of the “protocol sentence” suggested by Von Neurath:- “Otto's speech-thought at 3.17: At 3.16 a table was perceived by Otto”. I am putting myself in the utterer's (Otto's) place, and substituting a chair for the table.
My “speech-thought” would then be as follows:
The thoughts are actually compresent, but close temporal compresences can never be exhibited by speech, since it is inevitably a leisurely succession. But since we read from left to right, quasi-compresence can be visually expressed by vertical superposition, as it is in musical notation.
The simplest adverbial account of this experiencing would appear to be:
Since this is in visual symbols, i.e. in writing, each component could be replaced by a hieroglyph; “3.16” would require a diagram of a clock-face, with the hands in the appropriate position; “chair” could be replaced by a schematic drawing of the chair in question. The suffix “-ly” would follow each hieroglyph.
But in point of fact, “chairly” obviously won't do. Blind people experience “chairly”: at least, they use the word “chair” in unquestionably appropriate contexts. Thus “chairly” cannot express a modification of “visually”; we will therefore replace it by “chairshapely”.
“Visually 3.16-ly” could be analysed into more basic, geometrical types of visual concepts such as “circle”, “straight line” and “angle”. There is a direct structural connection between these and the function of temporal indication; we have to understand this connection before we are able to “tell the time”. But it would be a waste of time to try to analyse “chairshapely” in similar fashion; the geometry would be vague and indefinite, and there is no close structural correspondence between experiencing “chairshapely” and experiencing “chairly”. The latter could be expressed as the second part of the sequence
The use of the word “chair” to denote a purely visual mode of experiencing is due entirely to a frequent, fairly close temporal association between instances of “chairshapely” and those of “chairly”. These can never be “sensed” compresently, because experiencing “chairly” occurs in a part of the anatomy which is never within the field of vision.*
______________________
* Editor's note: Except in certain altered states of consciousness.
Nevertheless, the use of the word-sound “chair” indicates that the utterer must have been experiencing “tactually-chairly” just before the utterance; it is equally clear that this cannot have occurred at the intensity-level of sensation, but rather at the much lower level of reminiscence. The two have, however, been judged to be gap-indifferent in structure - like a sound and its echo. In point of fact if the chair is believed to be “real”, and not merely a picture, the intensity-level of the experiencing will be that appropriate to a proximately actualisable possibility, which is rather higher than that of mere “reminiscence”. We can now “score” the whole “experiencing” as follows:-
Actually, visually 3.16-ly
Chairshapely
Conjecturally soon, tactually dorsally
backsidely comfortably intensely
Wherever the word “chair” is uttered without reference to any specific instance of experience, it expresses the experiencing:-
Reminiscently Visually - chairshapely
Tactually dorsally
backsidely comfortably intensely
In this case the visual element will usually play an insignificant part in the “meaning” to which the utterer attaches the word-sound, since it has little intrinsic importance; nor is it currently functioning as a sign for the far more important tactile possibility.
The words “actually”, “conjecturally” and “reminiscently” here function like “marks of expression” in music. But the variations in the intensity of experiencing are enormous, ranging from the overpowering level of visceral and tactile pain to that of the remotest visual reminiscence, so that it is not possible to provide a more than rudimentary and quite inadequate expression of them, even in the adverbial language-mode. This, however, is notably superior to the subject-predicate style of ordinary written language, which virtually ignores them altogether.
We are often obliged to give them some expression in spoken language, because of the urgency of its signalling function. Urgent signalling expresses a high intensity-level of experience; it is always denotation, never reference. Our ordinary-language expression of the adverbial
“Actually tactually dorsally
backsidely comfortably-intensely”
would be “CHAIR!” The exclamation mark signifies the immediacy of this signalling function; nowadays, the large symbol <<!>>2 actually functions by itself as a road-sign, expressing “Optatively, ACTUALLY visually” or, in the vernacular: “Look out!”
______________________
2. When printed symbols are used for signalling, their extensive quantity, i.e. size, stands for the corresponding intensive quantity of the sound. But they may, on occasion, be too large to stand out sufficiently from their background. This point is of some importance in advertising. I am imitating C. I. Lewis's use of “French quotes” when I intend the reader to confine his attention to the visual aspect of a symbol.
Likewise, the meaning of the symbol <fire> is very different from that of the symbol <FIRE!>. The symbol <fire> merely signals the utterer's awareness of a certain distinctive series of gap-indifferent structures at the minimum intensity-level of reminiscence, whereas the symbol <FIRE!> signals an awareness at an intensity-level appropriate to the likelihood of its proximate actualisation as a “fortissimo” level of sensation. When read aloud, however, the contrast of dynamics will be no more than that between “piano” and “forte”.
In general we interpret our awareness of a written <word> as a faint, remote echo of the utterer's awareness, at the intensity-level of reminiscence, of the corresponding “word” sound. This is the primary “meaning” of every written <word>. This, however, is not the case where hieroglyphic writing is concerned; in this case the mode of “experiencing symbolically” is gap-indifferent with that of the mode expressed. It will only communicate successfully if the non-visual modes of experiencing - “connotations” - attached to the symbol by the utterer are gap-indifferent to those that the interpreters attached to the symbol - as, for example, the functional mode “chairly” that most of us are likely to attach to “visually-chairshapely”.† Such gap-indifference can generally be relied on within a region with a fairly uniform material culture, and dictionaries such as the Petit Larousse Illustré make use of this circumstance by using what are virtually hieroglyphs in order to reinforce their verbal definitions of words. Wittgenstein created something of a sensation when he suggested that the Tractatus could only be understood by those who had already thought along similar lines, but the remark is a truism, which applies to all verbal and symbolic communication. Even an ostensive definition of “chair” will be misunderstood unless the tactile “meaning” which the interpreter attaches to the appearance is gap-indifferent to that which the utterer intens to signal by drawing attention to it. It is only “chairshapely” that we believe to be experienced “conjointly”; “chairly” is experienced “severally”.
______________________
† Editor's note:According to Professor Fish, “Literature in the Reader: Affective Stylistics”, New Literary History 2 (Autumn 1970) pp.123-62, an informed reader brings to the text a network of linguistic, semantic, literary, and for our purposes, domestic, logical, philosophical, mathematical, scientific and spiritual experience that allows him to avoid a superficial or idiosyncratic response to what he reads - and it is a reader's obligation to become an informed reader! The typical reader, however, is not an informed reader. We may, therefore, add that the critical approach to any important group of texts begins with reading them straight through as many times as may be necessary in order to possess them in their totality, at which point the critical reader can legitimately begin to formulate a conceptual unity corresponding to the imaginative unity thus discerned in the texts.
Sometimes we use a “hieroglyph” to signal a mode of “experiencing visually” in cases where there is no non-visual connotation at all. The effectiveness of such expression appears to depend on the widespread attachment of æsthetic importance to certain modes of experiencing “visually-shapely” and “visually-colourly” just as it is attached to certain modes of experiencing “aurally-musically”. “Visually rectilinearly” and “visually circularly” seem to be of general importance, i.e., intrinsically “meaningful”, to a considerable portion of mankind. In these cases “appearance” is the sole function; the sign is not a “symbol”, but an “icon”, which replicates its “meaning”. The reference and denotation of the names for these shapes are therefore reliably gap-indifferent. This is how the notion of “self-evidence” originated - in the context of geometry. The so-called geometrical “proofs” served to draw attention to the more interesting and complex shapes that had been constructed out of the simpler ones, and recommended methods - sometimes alternative methods - whereby they could be replicated. Insistence on “ruler and compass” constructions reflected the comparative ease and reliability with which gap-indifferent instances of “visually-circularly” and “visually-recti-linearly” could be repeated. Of course, their æsthetic importance was reinforced by the discovery of their practical utility, but only in certain parts of the world. If Socrates' slave, in the Meno, had recently arrived from equatorial Africa, Socrates would never have been able to elicit his “previous knowledge” of geometrical “truth”; in those regions the “right angle” never left the vertical until the arrival of Arabs and Europeans, and linear measurement was virtually unknown.
The above analysis, rendered possible by the adverbial mode of expression, shows that gap-indifference between the denotation and reference of verbal expressions of visual experience can only be relied on in those few cases where no non-visual connotations are involved and where the basic modes of “visually-shapely” are of very general æsthetic importance. The attempt to extend “logical proof” to verbal expressions of anything other than connotation-free visual experiencing was always a futile exercise, since self-evident “axioms” are impossible; the denotative and connotative “meanings” of the terms employed cannot be guaranteed to be gap-indifferent. The “demonstrations” in Spinoza's Ethics furnish what is in effect a reductio ad absurdum of this procedure. “Self-evidence”, as the etymology of the term indicates, is strictly confined to the visual field.
The proper function of logic would appear to be disproof in cases where “self-evidence” has been mistakenly assumed, and can be shown to lead to absurdities. It should never be regarded as constructive, but is a valuable discipline for the purpose of criticising our verbal and symbolic constructions.
We might expect to discover cases of “self-exaudience” in the field of hearing - the equivalents of “self-evidence” in the field of vision. But the only “iconic” sounds are the onomatopœic words, and the range of their “meaning” is far too wide to provide a basis for “clear and distinct” concepts, although it clearly has its limits; we could hardly use the word-sound “BANG!” to denote a whisper with any hope that it would be “correctly interpreted”. Nevertheless, there is a form of “self-exaudience”, which is normally taken for granted; this is the structural gap-indifference between the utterer's speech-thought and his interpretation of the immediate “feedback” of the speech. The case of lapsus linguæ shows that this is not infallible, but we normally assume, not that an utterer necessarily “meant what he said”, but that he “said what he meant to say”. Speakers, qua utterers, do not attach a meaning to their speech; they attach a speech to their meaning. In the case of the initial utterance of every human being, namely, the wail of a newly born infant, we unanimously assume that the attachment happens “necessarily” and, furthermore, that the “meaning” is what we might later express by “Experiencing NASTILY!” Thus it provides the prototype of an “indubitable atomic statement”.* The “language” in which it is uttered is similarly interpreted by the whole of mankind, and the “predicate” is “atomic” in the sense of being undefinable in terms of anything else.
______________________
* Editor's note: Gilbert Ryle vigorously rejected this position when he reviewed Martin Heidegger's Sein und Zeit, and faulted its author precisely on account of his having implicitly propounded it - cf. Mind vol. 38, 1929, pp. 355-70. Dissenting from A. J. Ayer's exclusive appeal to sense-experience for verification, Ryle consistently held that no words used in true expressions are names in the sense in which names can be gesture-substitutes; judgment is transcendent because it always attains third-person truth.
The contrast between its simplicity and the complexity of our adverbial analysis of the meaning of the word “chair” - the mind of material expressed by the “basic statements” of physical science - is very striking. It illustrates the divergence between our requirements for signalling and description respectively. The expression:-
Conjecturally visually - chairshapely
tactually dorsally
backsidely comfortably intensely
cannot be spoken, and would in any case be much too lengthy for signalling purposes. Yet it expresses the structure of my mode of experiencing far more accurately than “chair”, and enables me to adjust the intensity-balance between the compresent elements according to the context in which the word “chair” is spoken. To achieve an even greater degree of “accuracy”, I would have to carry the analysis further still, and to provide spatio-temporal co-ordinates. The problem is analogous to that of describing an “instant” as defined by Whitehead in terms of an exhaustive totality of overlapping durations; a “complete” description would require the “naming” or “description” of every one of the durations. The “instant” can be signalled “quasi-instantaneously”, but its description is the very antithesis of “instantaneous”; from this point of view the structure of language inverts that of its “meaning”. This is particularly obvious in the case of spatio-temporal co-ordinates; every attempt at attaining greater accuracy involves an increase in the number of significant figures in their symbolic expressions. Obvious advantages to mankind have accrued from the physical scientists' efforts to analyse certain conjecturally “public” elements of experience into “atomic durations”, and the “logical atomists”, all of them fanatical admirers of physical science, seem to have been led by analogy to suppose that a description of an “atomic duration” would be an “atomic proposition”. It would be difficult to suggest a more erroneous assumption.
A “name” is a member of a gap-indifferent series of brief sounds, of which the primary function is to signal the distinctiveness of a member of a series of roughly gap-indifferent appearances. It is not, however, the appearance which provides the motive for signalling, but such of its functional connotations as are judged to be of widespread importance. The gap-indifference of the sound is therefore interpreted in terms of the functional connotations rather than the gap-indifference of the appearance. The important exceptions occur when the sole functional importance is a mode of “visually” or “aurally”, since the appearance itself is always in one of these modes. In the usual case, if and when our functional expectations are disappointed, we make remarks such as “John is not himself to-day” and it is clear that this negation is not conerned with the appearance. When the visual mode is of dominant importance, the mode of expression becomes rather different: “Those flowers are not real” may be taken as an example. Doubtless, “those are not real flowers” would be more appropriate, but in this context the most important mode of experiencing is “visually”, so that we signal it immediately and allow any matters concerning tactile connotations and mode of origin to be relegated to the more leisurely function of description; they are not “essential” to the immediate experience. This explains why we can use the term “flowers” to signal an experience which we proceed to describe as “unreal”. On the other hand, it would be nonsensical to say “That circle is not real”, since “circle” has no tactile or functional connotations. “That is not a real circle” is meaningful, signifying “visually-shapely, importantly differently from circularly”.
What a name signals is always a difference - a contrast between a current passage of aural or visual experience and its background. But the gap-indifference of instances of a name is attached to the actual indifference between our conjecture of the functional importance evoked by the appearance, and that of some gap-indifferent series of memories. Since the conjecture is likely to influence our course of action - in which we may include conceptual thinking - we are anxious to ensure that what is found “in the bottle” does not differ disastrously from the expectations aroused by the “label”; the latter always has to be as brief as possible in order to fulfil its signalling function, which is frequently urgent. The avoidance of such correspondence-failure is the only important function of a “logical” analysis of language; a very brief signal frequently has to call attention to a very complex compresence of conjectures, and we need to ensure that gap-indifferent signals do not evoke functional conjectures which differ importantly. When analysis is carried beyond this point it becomes mere pedantry, which may be amusing as a pastime but is otherwise quite useless. This point is well exemplified by the story of the learned professor whose wife caught him kissing the housemaid. To her indignant remonstrance:- “John, I am surprised at you!” he replied:- “No, my dear: it is I who am ‘surprised’; you are ‘astonished’.”
This is an extreme instance of the prevalent tendency of logicians to accord almost exclusive importance to the descriptive function of language, overlooking the fact that if it is “language”, it must have originated as an expressive signal emanating from an utterer. The “logical” point of view would not be seriously misrepresented by the sentence “In the beginning was the <word>”.
Translations of linguistic expressions into the “adverbio-musical” symbolism provide us with empirical analyses which reveal, far less inaccurately in in far greater detail, the structure of the experiencing which the utterer has been attempting to signal. To conclude this essay, I will apply this treatment to Russell's stalking-horse:- “Scott was the author of Waverley”.
He interpreted this as a description, equivalent to a compresence of three statements, namely:- “One man wrote Waverley: not more than one man wrote Waverley: that man was Scott”. But if I utter the sentence, I must be intending to signal my current experience, which bears hardly any resemblance to Russell's description, although Russell's interpretation can be derived from it by means of a complicated, backward inductive inference. Let us analyse what it actually signals, which turns out to be “information” of which the factuality is testable.
“Scott”, provided that we discount the homophony with “Scot”, signals nothing but the conjectural possibility of experiencing visually <SCOTT>-ly.
“Waverley” likewise signals the possibility of experiencing visually - <WAVERLEY>-ly, but also a distinctive gap-indifferent series of instances of experiencing “bookfunctionally”. “Was the author of” expresses a close spatio-temporal relationship between <SCOTT> and <WAVERLEY>, the former generally occurring below the latter on the front cover of the book and on the title page. The use of the singular form “author” signals that only one name is expected to appear in those positions. The use of “the”, as Russell correctly divined, combines an assertion with a negation, but its interpretation does not require a “unit class”, which is in fact a semantic monstrosity, since experiencing “uniquely” is experiencing “differently”, whereas experiencing “classly” is experiencing “gap-indifferently”.3 The negation, since it is a predicative compresence that is negated, can be expressed in the adverbial mode by a horizontal “stroke”. “Scott was the author of Waverley” can now be formulated as follows:-
Visually
un<SCOTT>-ly
Visually <WAVERLEY>-ly
Conjecturally, fairly soon Conjecturally, bookfunctionally, “Waverley”-ly
Visually <SCOTT>-ly.
Un<SCOTT> (signalled by “not Scott”) clearly denotes any symbol differing importantly from <SCOTT>. The information is, clearly, that the symbol <SCOTT> and no other symbol used for signalling a human being will accompany the symbol <WAVERLEY>, and perhaps also that if any other names does accompany it we have got hold of the wrong “Waverley”. Neither of the names denotes a “unit class”; the only ontological “unit” is an “atomic act of awareness”, which is far beyond the expressive range of any descriptive symbolism.
______________________
3. This “howler” corresponds precisely to that committed by the symbolic logicians when they substitute <v> for <and/or>, thereby abbreviating across the “stroke” which was inserted with precisely the intention of making this impossible. If <and/or> had been written <and> [but with those three superscript letters also underlined with, beneath that underlining and in subscript letters: 'or'], like <<±>>, one can hardly suppose that it would have occurred to them to do anything of the kind.
Earlier in this essay, I produced a tentative analysis of “tactually-chairly”, but I would quail before the prospect of attempting to apply a corresponding analysis to the complex, sophisticated “bookfunctionally”. I think that Carnap in his Logische Aufbau may have been contemplating something of the kind, and it is not at all surprising that he gave it up as a bad job. As we have already noted, the greater the descriptive accuracy of a symbolism, the more cumbersome it becomes, and adverbio-musical expression would be altogether too awkward for the purpose of communicating, which always involves signalling. How, for example, can we ever achieve a “True” description of our “meanings” in any symbolic medium which has to be spoken or read in a prescribed order? This requirement obliges us to use symbolic sequences to express temporal compresences of elements within our “experiencing”, and thereby utterly to misrepresent the temporal structure of what we are trying to express. We have become so accustomed to this practice that nobody, as far as I know, has previously drawn attention to the automatic falsification that it effects, although this becomes “tautologically” obvious as soon as it is pointed out.
It is this practice which makes it possible for us to express “unthinkable” compresences and then use the expressions for “descriptive” purposes, as in the case of the “unit class” considered above. We are obliged to use language, but we should always regard it with healthy suspicion, and never regard any symbolic description as “absolutely True”. This is no loss whatsoever; admittedly, false descriptions are often harmful on account of the “shock” which often accompanies the refutation of beliefs that they have engendered, but no additional “value” can be created by increasing the “truth” of a description beyond the point where the degree of its falsehood becomes unimportant. The ideal of “Truth” - with a capital T - is a bogus one. It is also pernicious, since the belief that its formulation is not only achievable, but has been achieved, is a potent source of fanaticism and intolerance. It appears wholly desirable to expose the absurdity of such a belief, and such an exposure is a major purpose of the present essay.
When we listen to a series of “verbal” utterances, or sequentially scan a collection of marks on paper which we believe to be a symbolic communication, we tend unreflectingly to assume that there must be some structural correspondence between the syntax of the symbols and that of their separate “meanings”.
This assumption is quite unjustified; indeed, when we “listen to”1 the situation from the utterer's point of view it becomes obvious that it is altogether mistaken. Almost every predicative description - even the simplest subject predicate phrase - containing two or more terms, radically falsifies the syntactical structure of its “meaning”. A simple example will serve to illustrate this point.
______________________
1. Not “look at”.
At the moment, looking out of the window, I can express my visual experience by means of the utterance “blue sky”, in full confidence that this will in fact serve to communicate my “meaning” to any English-speaking hearer. Nevertheless, the syntactical relation between the word-sounds is a gross falsification of the syntactical relation which prevails between blue and sky in actual experience.. The latter is an enduring, intimate temporal compresence of the two modes of visual experiencing, whereas the expression is a discrete succession of brief sounds, quite leisurely in tempo.
No alternative procedure is available; I have only one mouth, and even if I had some device for uttering both words at once, and of prolonging their utterance, the resulting sound would lack the “clarity and distinctness” required for the fulfilment of the signalling function which is primary to all language. I might be able partially to avoid this distortion were there a conventionally recognised abbreviation for “blue sky” - say, “bly” or “skue”. But the invention of such abbreviations, though sufficiently common, increases the voabulary and thereby places a greater strain on the memory of language-users, besides increasing the risks of misunderstanding, and we tend to avoid it except in cases where the “bundles” of predicates are of frequent occurrence and of immediate practical importance. In these cases single terms, rather than lengthy descriptions, are more useful for the purpose of signalling them.
In the current instance, the utterance “blue sky” is the best that I can achieve. However, since we are all constantly obliged to adopt this falsifying procedure, and have come to take it for granted, I can confidently assume that my hearers will reinterpret the expression in terms of an intimate temporal compresence of whatever invariant or gap-indifferent modes of experiencing they attach to “blue” and “sky”.
The fact that such re-interpretation does actually occur is demonstrated by the not infrequent cases in which it turns out to be impossible. For example, should I utter the phrase “a red white blue flag”, I will not expect my hearers to be able to “think” a sustained intimate compresence of “red”, “white”, “blue”. If, however, I insert and “and”, interpretation becomes quite easy. “Red white and blue” suggests a much looser degree of temporal compresence; “spatial” compresence within the visual field, according to Einstein, can be regarded as a rather intimate sub-species of the “temporal” variety. In this instance, the semantic function of the apparently conjunctive “and” is actually disjunctive; perhaps, indeed, it would be more accurate to say that a state of disjunction has always to be assumed before conjunction can be meaningfully effected.
It is interesting to contrast “a red white blue flag” with “a tall dark handsome man” which, like “blue sky”, is easily interpretable, since the disjunction is already taken for granted; “tall” and “dark” are sharply differentiated sub-modes of visual experience, while “handsome” is a mode of æsthetic appreciation. In the case the words provide what are virtually “qualitative co-ordinates” in empirically distinct “dimensions” of experience, whereas “red” “white” and “blue” are applicable to a single dimension, so that their “simultaneity” is unthinkable. If I write “a tall dark and handsome man”, my motive can only be the feeling that “and” improves the rhythm or euphony of the sound; it adds nothing to the meaning. Furthermore, alteration of the order of the words “tall” “dark” and “handsome” is unlikely to modify the interpretation of the phrase. On the other hand, there would be a natural tendency to differentiate between the interpretations of “a red white and blue flag” and “a blue red and white flag”, because we read from left to right, and if we imagine a tricolour, we wil expect the order of the colours to conform with that of the words.
Whenever an utterer does not suppose that conjoint predicates are compresently applicable, their order is nearly always semantically important. Should I announce that the 3 p.m. train from Waterloo stops at Southampton, Winchester and Basingstoke, I am uttering what may turn out to be a misleading falsehood, notwithstanding that I believe each of the seprate predications to be “true”. In this case, the falsehood will be “empirical”, whereas in the case of a statement such as “I arrived in Rome tomorrow” the falsehood will be “logical”, since I do not suppose that my readers will be able to think a “predicative compresence” of the past tense and the future temporal predicate.
I do not wish categorically to deny that some “Being” or other might be able to make sense of such expressions, as the saying goes, “sub specie æternitatis”. But as far as human beings are concerned, whenever we try to think the “simultaneity” of any pair of different modes of awareness within a single “dimension” of experience, the effort rapidly proves intolerable; we abandon it because it “costs us too much pains”, just as even the most talented athlete would speedily abandon any attempt to run a “one-minute mile”.
“Paradoxical” expressions are formulæ which are acceptable as descriptions of past events or event-sequences, but cannot be used to signal a present one. The simplest, most fundamental one is the vocal expression “F after C”, which is a “self-exaudient falsehood”,2 but is acceptable as a description of a former “predicative compresence” CF. The classical example “I am lying” is much more sophisticated, and indeed, if the thesis of the present essay is accepted, it is a tautology, since in no case can the syntactical structure of a subject-predicate statement correspond with that of its “meaning”. It is ridiculous to suppose that any “timeless truth” can possibly be formulated in a symbolising medium which has to be “scanned” at a finite tempo and in a prescribed order. The fundamental boundary for “ethical” purposes is not between “Truth” and “Falsehood”, but between “unimportant” and “important” falsehood.
______________________
2. Its visual equivalent is the “self-evidently” false [F to the right of C].
It is interesting to compare written language with another familiar form of visual symbolism, namely, the written notation of music, which actually contrives to exhibit the syntactical structure of its subject-matter.
The primary function of all music is that of signalling rather than description,3 since the symbols are used to convey instructions to the performers to go through certain motions calculated to reproduce the structure of the composer's sound-thoughts at the intensity-level of sensation. The syntactical relations between the “semantic” elements, which the composer has analysed out of his raw material, are invariably of great æsthetic importance, so that the notation has been designed to ensure that they will be replicated when the score is read, from left to right, at the tempo that the composer has indicated.
______________________
3. Only a few exceptionally gifted score-readers, such as the late Sir Donald Tovey, can actually memorise a piece of music directly from the score alone.
Vertical superposition of the symbols is therefore used to indicate intimate temporal compresences;4 in musical notation we would write, not (i):
the structure of which corresponds to that of normal speech, but (ii):
While (i) can be uttered by a “melodic” instrument, such as the human voice, this is impossible in the case of (ii); should the composer wish it to be uttered by purely melodic instruments, he will be obliged to write separate “parts” for them; the appearance of a “chord” in the part written for a melodic instrument is a “logical falsehood”, since it cannot be “interpreted”.
___________________
4. These must always fall short of “absolute” simultaneity, since musical “pitch”, which is what we call our awareness of a certain range of invariant “frequencies”, can hardly be regarded as existing at an “instant”.
In any event, a chordal compresence lacks the “clarity and distinctness“ of a single sound; in the case of purely musical sounds, this does not matter, since they do not serve to signal any further “meanings”, and the composer has presumably regarded their compresence as æsthetically superior to their mere succession. But in the case of spoken words, each of them is intended to function as a signal which stands for some mode of experience outside the sphere of language altogether, and this function requires that they shall be as clear and distinct as possible.
The contrast is illustrated very well by operatic ensembles in which each of the characters is uttering different words; the ensemble of the musical sounds may be very beautiful, but the “sense” of the text is blurred and confused. We can usually judge the nature of the situation, because the words, as a rule, repeat passages which the singers have already uttered “solo”. When this is not the case, such ensembles are, semantically speaking, practically null and void.
It is clearly impossible to exhibit a relation of intimate temporal compresence by means of any symbolism in which the semantic units have to be uttered and read in a leisurely linear succession. In the case of written music, the extra dimension provided by the visual field makes it possible to confine the use of the horizontal dimension to the indication of temporal succession. The same technique is used for the graphic representation of change in general; we read from left to right, and the order of the changes indicated corresponds to the order of our reading. There is therefore no need for any vocabulary for the indication of temporal relations; the paradox “F after C” could only be “actualised” by reading the notation backwards. A proposition of Gödelian type - not “provable” within the symbolism - has to be accepted as “true” before any collection of marks on paper can even become a “symbolic expression”. This is: that the collection has been uttered, and is intended to be “scanned”, at a finite tempo and in a certain prescribed order. In the case of Indo-European languages, the scanning order is, comparatively rapidly from left to right, and, more slowly, from the top to the bottom of the page. Mathematical and logical symbolisms follow this pattern; even Semites and Chinese have adopted it for these purposes, although it does not correspond to the order in which they scan their own written languages. I am ignorant as to the order actually adopted by the Chinese, but it is evident that some such order must exist, since the “language” can be read aloud.
It is clear that these conditions are bound to impose radical limitations on the expressive powers of all symbolism. Its syntactical structure is so coarse-textured that it cannot possibly achieve more than an extremely remote approximation to that of any intimate temporal relationship. Its signals can be used to draw attention to the occurrence of such relationships, in the hope that the modes of importance attached to them by the hearers will not differ excessively from those to which the utterer has attached them. If this condition is fulfilled, the language will obviously “work” in a pragmatic sense. But the notion that it can ever achieve an accurate description of anything more delicate than certain comparatively coarse-textured elements of phenomenal experience is so absurd that it is not altogether easy to understand how the notion of “Absolute Truth” ever came into existence, and its apothetosis as an intellectual “Ideal” is completely fantastic.
It is the defectiveness of our symbolism which requires the disciplines of “logic” in order to ensure that its deficiencies do not betray us into using what are in fact “logical falsehoods” for signalling purposes. In view of our general practice of using sequences for the purpose of signalling compresences, it is especially important to examine every abbreviation in order to make sure that it does not conceal an important inconsistency.
Here some of the most eminent logicians have been clearly at fault. Take, for example, Leibniz's term “indiscernibles”. This embodies a logical contradiction, since the use of the plural implies that discernment has actually occurred; a state of disjunction must be assumed before conjunction can be effected. The final “s” could be replaced by the use of the predicative adjective “plural”, which must assuredly be applicable to any entity for which a “universal” can stand. Russell considered that Leibniz's belief:- “There are many monads” constituted an inconsistency in his logical system, because it was not of subject-predicate form. But the belief could equally well have been expressed by means of the signal “Monads!”, and this in turn could be correctly reformulated as “Plural Monad!”, just as “Fires!” could be reformulated as “Plural Fire!” The substitution of the plural suffix for the initial adjective is merely an abbreviation adopted for the purpose of conciseness, always a desideratum for signalling purposes.
Whenever an abbreviation is explicitly invented for this reason, the synonymity of the full and abbreviated forms can be unconditionally accepted. In cases such as the use of “Comintern” for “Communist International” this is obvious, but it applies also to the use of a word like “stallion” for the purpose of signalling a sexually active male horse. Matters are quite otherwise in cases where we encounter the signal long before we ever attempt to define it; one such case, cited by Quine, concerns the “synonymity” of “brother” and “male sibling”. Here the signal word is not “purpose-built”, so that we feel no serious objection to the use made of it, for example, by trade unionists or nonconformists or, indeed, in any context where its meaning does not differ excessively from that to which we originally learned to attach it.
There is quite a strong case for applying the distinction synthetic /analytic to terms rather than propositions. Concepts such as that expressed by “stallion” have been generated by synthesis, whereas those expressed by terms such as “crimson” can only have been generated by analysis carried out within the field of an existing concept - in this case, “red”. A “synthetic” concept is created when importance is attached to a de facto overlap of the extensions of its parent concepts, whereas an “analytic” one results from an original attachment of importance to a qualitative boundary within the field of a single parent-concept. A “synthetic” term is conjoined to its several parents exclusively by relations of similarity, whereas an “analytic” one is generated from a single parent by a process of differentiation, or “multiplication by fission”. The latter is never an abbreviation and, therefore, can never be adequately defined by means of any collection of existing terms.
Now conceptual synthesis has obviously been operative for a long time before we begin to speak, and the concrete nouns which play a large part in our primitive vocabulary are brief, highly synthetic terms used for signalling “bundles” of phenomena to which we attach importance either for æsthetic or practical reasons. It is only when we attempt to compare and contrast them that we begin to take note of the individual “qualities” that they embody, and to invent or learn words in order to express them. In musical metaphor, we can say that our phenomenal material consists mainly of “chords”, and that the individual “notes” only become important when we wish to replicate them. Every composer is in much the same case. He can perfectly well enjoy his private experience without undertaking the processes of analysis and re-synthesis involved in “composition”, but, wishing to enhance his own enjoyment by sharing it, he accepts the labour in order to replicate it at the intensity level of sense-perception, so that it can be communicated to others.
Our usual motive for signalling the phenomena symbolised by concrete nouns is not primarily a desire to share our enjoyments, but rather to co-operate with others in avoiding disenjoyments, especially those which we confidently believe to be quasi-universal concomitants of certain types of contact with surfaces. Warnings against the possibility of such contacts are the most urgent kind of signal. These have no counterpart in music which, unlike language, is exclusively aimed at the direct enhancement of enjoyment.
For this reason, there is little parallelism between the “semantics” of music and that of ordinary written language; it is precisely the absence of semantic complexity in written musical notation which renders it so admirable as a basis for criticising the structure of ordinary language. It does, however, contrive to exhibit relative degrees of difference within one semantic “dimension”, namely, that of “pitch”. The principle is simple; the difference in vertical position on the stave between two note-symbols varies directly with our empirical judgment of the difference between the pitches of their “meanings”.
Such a method of differentiation is ruled out in ordinary language, on account of its unidimensional, linear character. Musical notation seems to have been a partial anticipation of the methods of Descartes' co-ordinate geometry, and it is interesting to note that when these are used for the graphic representation of change, the variable “time” is always plotted horizontally, differences of quantity being represented by differences of vertical position.
It should by now be clear that the structure of ordinary language must always be a hotch-potch, bearing little relation to that of what we attempt to use it to describe. The “rules” of grammar and syntax are not constructive and “a priori”, but prohibitive and “a posteriori”. They represent the activity of “human-natural” selection, based on a certain community of æsthetic judgment, and their purpose is to eliminate from language sequences which have led to the expression either of the uninterpretable or, in contexts where it might be “instrumentally” dangerous, of the ambiguous. In the context of poetry, ambiguity is often a positive virtue, which explains why a considerable relaxation of the rules of grammar and syntax is generally acceptable. Even poets, however, tend to avoid formulating the uninterpretable, since this is bound to occasion a slightly unpleasant “shock”, especially to anybody who possesses a sensitive ear for language. Uninterpretability is never exclusively a matter of surface-structure; it results from an irreconcilable incompatibility between the surface-structure of the language - which is to say, the order of the words and the implicit degree of temporal compresence between their “meanings” in its normal interpretation - and the implicit syntactical relations entailed by the “meanings” of the terms employed.
“Logic” is a schematic exposition of the rules of grammar and syntax. In the present essay it is maintained that the “linear” structure of ordinary language, which has also been adopted by “symbolic” logic, renders it virtually ineffective for this purpose, and suggests that an appreciable improvement can be achieved by using a two-dimensional “musical” style of symbolisation. “Perfection” is out of the question in any case; direct experience is so wildly “polyphonic” that attempts to describe it by means of a “melodic” line must always be tentative and wide of the mark. Nevertheless, criticism from the “musical” standpoint can help us to eliminate many serious errors, including some which seem to have become embodied in traditional logic itself.
The expression “I am lying” can only be regarded as “paradoxical” on the assumption that a judgment of the “truth” of an expression can be made during the period of the utterance, or during the reading of the sequence of written symbols which enables us to replicate it.
It is clearly impossible to assert the “truth” of any assertion until it is considered to have been completed; for example “I am lying, without the final inverted comma, might simply be the beginning of “I am lying down”.1 Furthermore, a short but appreciable duration will be required for its interpretation, that is to say, for arriving at a conjecture as to the nature of the something-other-than-itself which it is presumably intended to communicate. A judgment as to its truth is eventually applied, not to the expression itself, but to an echo of it which follows it so closely that we regard it as structurally identical with the original.2
______________________
1. A “logical” falsehood can often be detected during the course of an utterance. For example, if I should make the assertion: “Tomorrow, I arrived at Rome”, the falsehood would be obvious before the words “at Rome” were considered.
2. This is an application of Russell's “Principle of Quasi-Permanence”.
Thanks to the immediate feedback resulting from the mutual proximity of the organs of speech and hearing, the utterer himself is always the first hearer of his own utterance; he may even, on occasion, be the only one.* He thus has an immediate opportunity of interpreting his own utterance, and, since he is bound to have an extremely recent memory of its intended meaning, he is in a uniquely privileged position to judge whether his interpretation of it corresponds with what he intended it to express. Now there are obvious instances when an utterer can make an immediate and authoritative judgment of the falsehood of his own utterance, in which case, “I am lying” will be as “true” as it is possible for any statement to be. Such instances are familiar in the shape of “lapsus linguæ”. When the utterer has meant, and presumably still means, to utter the words “I am telling the truth”, but actually hears “I am lying”, then his utterance has indeed falsified his intended meaning. Similarly, should he say “This statement is false” when his intention is to say “This bottom is false”, he is telling the truth; “this statement” has falsified his meaning.3
______________________
* Editor's note: Either literally or metaphorically. If the utterance is felt to be importantly significant, such a circumstance may be experienced as dereliction in a sea of loneliness.
3. Cf. Tarski:- “This statement is true” is true if, and only if “this statement” is true, i.e., is a correct expression of the intended meaning. It is assuredly false if “this statement” was not the intended utterance.
In such cases of unintentional lying the utterer will ordinarily take steps to rectify the situation as soon as he becomes aware of the failure of correspondence. In cases of intentional lying the correspondence failure is rather different; it is between the utterer's interpretation of his own expression and something which was uppermost in his mind slightly prior to his initiation of the utterance. In order to utter the lie, he has had to counteract his habitual propensity for expressing “spontaneously” what would be easiest to express, and discover or invent something which differs importantly in some rather precise respect from the thought of which he is most vividly aware. He then has to maintain the two thoughts in very close compresence for whatever period is needed to express the second one.
We can perhaps most advantageously regard such a compresence as an “oscillation” between the two; if its “amplitude”, i.e., the “felt” difference between them, is very considerable, a considerable effort will be needed in order to maintain it. The physiological concomitants of such efforts actually produce “psycho-galvanic” reactions from machines of the “lie-detector” type.
The “present” tense of any verb embodied in an assertion can never be regarded as “historic”, i.e., theoretically “instantaneous”, since a certain duration is required for its utterance and interpretation.4 The speciousness of the “present” is very variable; “I am lying” would express a continuing mode of experience with a minimum duration which embraces the utterance and interpretation of the expression, whereas “I lie” or “I am a liar” expresses the habitual recurrence of such modes over a much longer period. For the interpreter of any expression, the period includes the duration
of the utterance and a brief sequel, but for the utterer, it must also be partly antecedent to the utterance. Thus “I am lying” will express the utterer's state, not merely during the utterance, but for a short time before and after it.
______________________
4. This includes the copula, “is”.
A description of the utterer's psychological state during the utterance of a lie has been given above. If it is accepted as even approximately correct, it is clear that “I am telling the truth” can be true when, and only when, an utterance is spontaneous. But “I am lying” can be a spontaneous signal of the psychological state which accompanies the utterance of an intentional falsehood; “lying” is not, indeed, a predicate of the statement, but of the utterer!
Other versions of the “Liar” can be disposed of quite easily. Epimenides the Cretan, who is alleged to have asserted that “No Cretan ever tells the truth”, may have been guilty of a lapsus linguæ - “ever” for “never” is an obvious possibility - or, alternatively, he may, by his utterance, be falsifying his own statement by terminating the “specious present” during which it may have invariably corresponded with the facts. The indirect forms of the paradox are still easier to resolve, since the two assertions which are supposed to add up to an incompatibility are not uttered simultaneouslly. Here is a version given by Popper:5
The statement (a), as it stands, is meaningless, since no assertion can be “true” or “false” before it comes into being:6 We can make it meaningful by substituting “will be” for “is”, but in this case it merely describes the current intention of the utterer, and he has plenty of time to change his mind before he utters (b). Alternatively, it may misdescribe is intention, either intentionally or unintentionally, in which case (b) is clearly true.
______________________
5. Conjectures and Refutations, ch. 14.
6. Cf the classical problem of the Aristotelian sea-battle.
In point of fact, many expressions which are in everyday use are grossly paradoxical; when this is pointed out, their utterers generally reaction in much the same way as they would to a “lapsus linguæ”; they have uttered something which differs importantly from their intended meaning.
The most important case concerns the use of the word “after” (or the verb “follows”) to express a temporal relation. The spoken utterance “F after C” reverses the syntax of its intended meaning, since F is spoken before C. This has its written counterpart in “F is to the right of C”…
In Hebrew, it would be the semantic equivalent of “F is to the left of C”. Symbols function as mnemonics for word-sounds, which in turn function as mnemonics for complex acts of awareness which are, for the most part, unconcerned with language. However, in the case of those who can read and write, these complexes include the visual appearance of their own symbolic mnemonics, which become important whenever we discuss the functioning of language. In the examples given above, the symbols F and C are mnemonics for the sounds “eff” and “see”; these in turn can serve as mnemonics for musical notes, but in the present context, the most prominent elements in their “meanings” will most likely be the appearance of their own visual mnemonics.
Now a word-sequence, when spoken or read, is an event-sequence, and its order is unalterable. But we can arrange the mnemonics for separate parts of the sequence in any order we like.7 The temporal relation “after” cannot be ostensively defined, since there is no “backward” sequence of events, but we can reverse the order in which we display the “facts”, which express memories of them. When we retranslate the visual symbols into their audible equivalents, they always have to be scanned in a prescribed order, which will, of course, be that in which the words are spoken. Unfortunately language, unlike musical symbolism, cannot possibly exhibit the temporal structure of its subject matter, and uses various indirect expedients, such as affixing temporal predicates to the mnemonics for events, and relating them either by temporal prepositions or verb-tenses, in order to try to express it. This is how it becomes possible for the spatio-temporal order in which we speak or read the mnemonics to reverse that of the events which we are using them to describe.
______________________
7. The sequence of events is McTaggart's “A” series; the conventionally agreed order in which we arrange our memories of them, the “B” series. The relation-word “after” can only be used to conjoin members of the “B” series.
It will, I think, be generally admitted that the sequence CF is less incorrectly expressed by “C before F” than by “F after C”, although the two expressions would usually be considered to be synonymous. The structural correspondence between event-sequences and expressions could be used to provide an ostensive definition of “before”; ostensive definition of the dyadic relation “after” is impossible! Similarly, “C to the left of F” can be used to provide an ostensive definition of “to the left of”, when read in the customary direction, but this does not work in the case of… “F to the right of C”. When judged by Tarski's formula for correspondence-truth, the latter must assuredly be false, since “F is to the right of C” is true if, and only, F is to the right of C.
We obviously require an explanation of how it comes about that expressions which are “logically false” are nevertheless in common use. This is not at all difficult; when we utter the “true” formula, “C before F”, the utterance of “see” and “eff” can be quasi-simultaneous with the corresponding acts of awareness at the intensity level of sensation, but when we utter “F after C”, if “eff” is simultaneous with a sensation, “see” can only be simultaneous with a memory, which is not “ostensible”. We never actually “refer”, in the sense of speaking or thinking “backwards”. So-called “reference” is a transition of awareness from a sensation to a memory. We can hardly be mistaken about which is which, on account of the enormous difference between the intensities of the two kinds of experience.8
______________________
8. “Reference” may also be a transition of awareness from one memory to another, to which an “earlier” temporal predicate is conjecturally assigned. Mistaken assignations are quite easy to make, since the intensity of a memory-echo seems to vary, not only inversely with its remoteness, but also directly with the importance of the experience memorised.
Now written symbols, with which logic is primarily concerned, provide no indication of the intensity with which their utterer has experienced whatever he has used them to express; a structurally gap-indifferent series of symbols is attached to a structurally gap-indifferent series of acts of awareness, irrespective of any differences in their intensity. From a logical point of view, therefore, there is no difference between an experience and its memory-echo.
From the utterer's point of view, however, the “same” word may express a truth or a falsehood. For example, the “logical” meanings of the symbol <fire> and of the word-sound “fire” do not differ from that of the word-sound symbolised by <<“FIRE!!”>>, but from the utterer's point of view, the difference between them is of the “uttermost” importance. He can hardly judge his own utterance of the isolated symbol <fire> or of the word-sound “fire” to be “false” - unless, perchance he has intended to utter, say, “water” - but he will be well aware that “FIRE!!” is false when, immediately before the utterance, the intensity-level of his awareness of the stucture was no higher than that appropriate to indefinitely remote memory. Thus a single spoken word can express a falsehood; in the case of a written word, this is only possible when we encounter it in a context in which we have learned to regard it as a signal, and its signal-function will in any case have been vocally explicated. Under these circumstances it functions as a mnemonic for a shout. The linguistic symbol <<!>> is actually used by itself as a road-sign, simply to call for intensified awareness on the part of motorists.
On the view adopted here, the “past” is almost as unreal as the “future”, being manufactured out of the continuing present by a process of “backward induction”. We are all the time aware of the rapid fading of sensations into memory-echoes, and tend to assume that every memory-like experience must be the echo of some counterpart at the intensity-level of sensation. Since linguistic communication occurs at the intensity-levels of visual and aural sensation, Hume's empiricist principle, “No idea without antecedent impression”, is doubtless apllicable to everything that we can communicate by word or symbol. But it by no means follows that it is applicable to everything that we can experience; the rhythmic texture of our symbolism, especially that of speech, is very coarse, and there is reason to suppose that much - perhaps most - of our experience is bound to slip through its meshes. “Ineffability” is probably the rule rather than the exception, and is mainly due to the expressive inadequacies of our “effing” apparatus, the powers of which have been grossly overestimated.
In this essay, I propose to discuss the use of these words as predicates, whether applied to names or descriptions. There is a strong tendency, in the writings of philosophers such as Russell, to treat them as synonymous, but this seems to be a mistaken policy. Their etymologies are altogether different, and two such distinctive words can be very useful for the purpose of expressing two distinct modes of experience.
“Existere” is clearly derived from “ex” (out of) and “stare” (to stand). It should therefore be in order to attribute “existence” to anything which stands out from the general background of experience, even though the “existent” manifests itself privately, within a single “biography”. Furthermore, we do not interpret any vocal sound as “linguistic” unless we assume that it is the sequel to an attempt by its utterer to express some distinctive mode of his own experience; this I call the “significant” of the expression, by analogy with Peirce's “interpretant”. It is clear that our acceptance of any utterance as “linguistic” implies our assumption of the “existence” of its significant, even when the utterer's mode of expression is such that we are quite unable to conjecture its nature.
It now becomes clear that “existence” is otiose as a predicate, because we have already accepted it in the case of everything that is verbally expressed. It therefore effects no definition whatsoever; “sine negatione non est determinatio”, as Spinoza might have said, had it occurred to him to express his famous dictum in modus tollens.
The limits of “reality” are far more circumscribed. The word is clearly derived from “res”, which is usually translated as “thing” but is more frequently used to express “business” or “affairs”, as in “republic” and “in re so-and-so”. It would seem to indicate an assumption of the common importance of its reference to the communicants. Its usual translation reflects the common importance of “things”, i.e. solid objects, to every member of the human race.*
______________________
* Editor's note: Two points: (1) Bernard Lonergan's Insight - A Study of Human Under-standing (revised students' edition, Longmans 1958; posthumous critical edition, University of Toronto Press 1992) contains an entire Chapter explaining the distinction between “things” and “bodies”; (2) For Aquinas and other mediæval Latin writers “res” (i.e. “thing”) normally prescinds entirely from the question as to whether the “thing” under discussion actually ex-sists or not, connoting rather its concrete intrinsic possibility of so doing - not all reality is actual. Solomon's existents approximate to Lonergan's & Aquinas's realities and vice-versa.
The best way of defining the appropriate field of application of any predicate is to attempt to establish a dividing line between cases in which it is appropriate and those in which it is not.
Now everything that we can communicate with confident expectation of general success clearly comes within the category of “real”, so that we do best to begin by examining “significants” which will be generally regarded as closest to complete “unreality”, that is to say, those expressed by words which do not succeed in communicating at all.
For examples we have to turn to the works of writers such as Edward Lear and Lewis Carroll, and even here utter “meaninglessness” is rare. But “runcible” as the attribute of a spoon and “vorpal” of a blade would appear to qualify. “Snark” and “boojum” likewise have excellent claims to “unreality”, although the context establishes that Carroll's significants must have included modes of visual experience.
But when we come to “Jabberwocks”, “Borogoves”, “frumious” and “slithy” we are already on far less certain ground, while “chortle” from the same poem has even got into the dictionary. The “animals” have only attained visual “reality”, but Humpty Dumpty's elucidations of the significants of “frumious” and “slithy” are adequate enough to permit the use of those words, where appropriate, for descriptive purposes, while “chortle” clearly satisfied a need for an onomatopœic word to express a distinctive type of sound to which it was not yet possible to make adequate reference.
The “animals” for readers of Alice must be regarded as just as “real” as “circles” or “straight lines”, since mention of the words immediately evokes the conjecture of the possible actualisation of their significants at the intensity-level of visual sensation, which is the highest expected, in the shape of Tenniel's illustrations. Nevertheless, the words would have remained as unreal as “runcible” and “snark” until Tenniel illustrated them or Humpty Dumpty described them; only “chortle” could speak for itself. Humpty Dumpty defined them in terms which belong unmistakably to the field of “reality”, while Tenniel for his illustrations employed techniques which usually serve to represent visual sensations which are interpreted as signs for solid bodies. This is why the animals still remain in an important sense “unreal”; the descriptive technique employed is one which is usually intended to evoke conjectures of the possibility of actualising an experience at the highest intensity-level, which is that of tactile sensation. In this case it has been used to express a mode of experience at the lowest intensity level, namely, that of indefinitely remote conjecture. Visual sensations are only accepted as signs for “realities” when their conjectured interpretants are such that their possible actualisation, which must not be indefinitely remote, will be more intensely felt than the sign-sensations themselves. A marginal case of “reality” occurs in the case of copies of works of art. The original is generally described as “real” because it is assumed that it will evoke feelings of greater intensity than those evoked by the copy, although the difference is often very slight, and the assumption may even, on occasion, be unjustified.
The same considerations apply in the case of verbal signs, but word-sounds are of so little intrinsic importance that their significants can seldom be “unreal” in this sense. “Runcible” and “Vorpal” qualify, since their importance seems to depend entirely on their euphoniousness within the context in which they occur.† I can think of nothing of this kind in prose. The “unreality” attributed to the significant of an expression containing more than one word depends on its failure to evoke any conjecture, at the intensity level appropriate to the possibility of its proximate actualisation, of the simultaneous experience of the references of the individual words at the highest level of intensity appropriate to them.
______________________
† Editor's note: Since J. D. Solomon penned these words a double-sided, double-density 5¼" computer-disk has come onto the market, containing a “Vorpal Program” that considerably enhances the performance of the Commodore-64 electronic computer's otherwise notoriously sluggish disk-drive and can, indeed, frequently detect, diagnose and quasi-automatically correct defects in its mode of operation.
Thus “golden mountains” remain unreal, in much the same way as “jabberwocks”, until such time as a “golden” visual sensation evokes the conjecture of the proximate possibility of actualising a “mountainous” tactile-muscular interpretant at the full intensity-level of sensation. The “unreality” is only empirical, since there is no contradiction within the verbal combination. When we come to consider unrealities such as that expressed by “square circles”, matters are rather different. This is a logical contradiction, since the contrast between the significants of the separate words has clearly played a part in establishing them as “meaningful”; “having corners” and “having no corners” is a difference to which we have surely attached importance before we ever learned to speak. The expression can therefore evoke no conjecture of experiencing both of their significants “here-now” at the level of visual sensation; any attempt to imagine it results in a fruitless and somewhat painful psychological effort.
Nevertheless, the “significant” can still be considered to “exist” at the very low intensity-level of indefinitely remote conjecture; nothing which is verbally expressible can belong to a “null” class. The psychological difficulty of making a rapid transition between the two significants at the very low intensity-level of indefinitely remote conjecture is so much less that it becomes negligible.
The status of significants such as that of “square circle” presents something of a problem to anybody who wishes to maintain an empiricist hypothesis. The significants of the constituent parts can certainly be regarded as memories of sensations, but how about the structure which results from their immediate juxtaposition?
Similar problems arise out of all “creative” thinking. It is not clear whether acts of awareness of new structures, arising from a combination of experiences at the very low-intensity-level of indefinitely remote conjecture, should be regarded as “impressions”. The problem is especially acute in cases such as that of “square circles”, since it is virtually certain that the compound structure is not a memory of sense-experience, but it equally concerns the significants of all fiction. At what level of intensity, for example, did Shakespeare first experience Hamlet?
In order to maintain a consistently empiricist position, we will have to admit that Hamlet originated in the structure of an experience which happened to Shakespeare, but at the very low1 intensity-level of indefinitely remote conjecture. In a sense, Shakespeare should not be regarded as its creator, but rather as its translator into the medium of words. Artistic and intellectual “creation” thus becomes a combination of selection and expression by human beings of outstanding sensitivity, who seek to communicate the beauty of certain structures experienced by them at many intensity-levels, including that of indefinitely remote conjecture. It would appear that the fine-structure of “impressions” experienced at this level can be must subtler than that of those experienced at the ordinary level of sense-perception. This view seems quite tenable and is in accordance with the testimony of many famous writers, artists and composers. Such structures must clearly be expressed at the level of visual or aural sensation before they can be communicated; once this has been done, there is a sense in which they can be regarded as “real”, since subsequent use of their names for reference purposes will evoke proximate conjectures of the possibility of actualising their expressions at the level of visual or aural sensation.
______________________
1. I have revised my opinion on this point, as far as the author is concerned. See the postscript to this essay.
Yet in another sense they remain “unreal”, since they express significants which were not orginally experienced at the “normal” level of sensation, but at the much lower one of indefinitely remote conjecture.2 It is this criterion which separates “fiction” from “fact” and “non-representative” from “representative” in art. Music is the only art which is virtually free from this distinction; apart from occasional attempts to imitate non-musical sounds, it is pure “fiction”, and evokes no high-intensity conjectures whatsoever. It can be unreservedly accepted as “real” at the intensity level of aural sensation.
______________________
2. This only applies to the reader, not to the author.
It is clear that the distinction between “real” and “unreal” stems from the circumstance that visual and aural sensations, which furnish us with our media of communication, occupy an intensity-level intermediate between those of tactile-functional sensations on the one hand, and indefinitely remote conjectures on the other. The primitive and habitual use of visual and, to a lesser extent, of aural sensations, is a sign which evokes conjectures of the possibility of tactile and functional ones of a much higher level of intensity and, therefore, much more than important than the visual and aural ones themselves. We learn this habit in infancy, and retain it throughout life, especially in the case of visual sensations.
As we grow up, however, our æsthetic sensibilities increase, and we begin to attach intrinsic importance to variations within the fields of visual and aural sensation and, later still, within the subtle, low-intensity field of memory-conjecture. Language, which has developed out of the human urge to co-operate in the tactile and functional fields, is too coarse-textured to do justice to the finer elements of experience; attempts at their verbal expression resemble those of a photographer who tries to take a very detailed photograph, using film with a coarse-grained emulsion. This is bound to fail as soon as the grain of the detail required is finer than the grain of the emulsion, and the detail therefore remains “unreal” from the point of view of anyone examining the photograph.
It is now possible to put forward a formal definition of the “unreal”, as follows:-
“The significant of a verbal (or symbolic) expression is ‘unreal’ to its hearers (or readers) in so far as any of the conjectures, which the expression evokes, of the possible actualisation of distinctive structures at the full intensity appropriate to the mode or modes of sensation and/or thought involved, are not experienced by them at any intensity-level above that of indefinite remoteness.”
The significants of expressions in “nonsensical” and “metaphysical” language remain “unreal” for their hearers as long as these conjecture that nobody is able to find interpretants for them which he, in his turn, can re-express in “real” English. This is what distinguishes them from, say, the significants of expressions in Chinese or Higher Mathematics.
It is clear that unreality varies in degree. The significant of a word can be real in some contexts but unreal in others. In the following series of expressions the significant expressed by the word “fire” progressively approaches “unreality”:
Of these, only (i) expresses a direct tactile sensation, or the likelihood of the immediate actualisation of the conjecture of one. In the cases of (ii), (iii) and (iv) prospects of actualisation become increasingly remote, and the degree of “unreality” of the fires at Timbuktu can be compared with that of “stars” that are said to be billions of light-years away. Actualisation of such conjectures, if it ever occurs, can only be in the distant future, and cannot in any case “verify” the present ones.
Only in the case of (v) is there no prospect of tactile actualisation; the degree of unreality in this case is similar to that of the Jabberwock, since, for readers of an illustrated version of A Christmas Carol the “fire” may be actualisable as “that!” at the level of visual sensation.
If any statement is intended to be actualised - i.e., verified or refuted - by any persons other than the utterer himself, it is clear that their interpretants must not differ importantly from his significant. The utterer can hardly assert its “truth” unless he believes that this will be the case.
There is bound to be an element of uncertainty here, since an interpretant of a verbal expression is no more than the hearer's conjecture as to the nature of the significant that the utterer has “meant” it to express.
In the case of commands, requests and immediate warnings, correspondence between significant and interpretant is either corroborated, or refuted, by the appropriateness of the immediate response, followed in turn by signs of approval, or otherwise, on the part of the utterer. But in the case of information, actualisation by the hearers may be postponed indefinitely; an utterer who asserts that his statement is “true” is merely expressing a confident conjecture, based, no doubt, on his own experience, that if and when any of his hearers chooses to actualise it, the actuality will not differ importantly from the expectations evoked by their interpretants of the statement. The utterer therefore cannot validly assert the “truth” of a statement unless it expresses a significant which he has experienced wholly at an intensity-level higher than that of indefinitely remote conjecture; it must have some degree of reality. Once it has been uttered, the statement itself immediately achieves reality at the level of aural sensation, but not at any higher level.
We can assert “truly”, though Lewis Carroll could not, that “Jabberwocks will be found in Looking-glass Land”, provided that we do not look for these outside the pages of Alice, or perhaps its translation into a cinematographic form of expression - Jabberwock-appearance is visual “reality”. Unless this audio-visual level of experience is accepted as “real”, there can be no “correct” answer, for example, to any question in an examination paper on English Literature!
Outside literature, fictitious beings can exist, but cannot truly be asserted to be “real”. “The present King of France exists” is a tautology provided that “existence” covers the lowest intensity-level of experience, which is that of indefinitely remote conjecture. On the other hand “The present King of France is real” is false, since it is likely to be interpreted in terms of the proximate possibility of actualising certain fairly well-defined types of visual and tactile sensation, compresently with sequences of “kingly” function.
The witticism suggested by Professor Strawson: “The present King of France is the only wise ruler in Europe”, depends for its effectiveness on the assumption that the hearers will take his “unreality” for granted; their equation of the “unreal” with a member of a supposedly “real” class, namely: “wise rulers in Europe”, is, like most verbal humour, the intentional expression of an unimportant falsehood. It could be paralleled in mathematics by the intentional equation of a “real” with an “imaginary” expression - but mathematicians within the confines of their subject are not supposed to indulge in this kind of “joke”, since to them all falsehoods are desperately serious.
This latter view is taken by most philosophers who approach their subject from the side of science and of mathematics, which provides it with its “language”. Their ideal of “Truth” theoretically requires “perfect” communication, that is to say, the postulation of the “identity” or at least the “quasi-identity” of the significant and all interpretants of every symbolic expression. Such quasi-identity can only be tested provided that their statements are such as can be repeatedly actualised, so that their methodology excludes the whole of the “unreal” from their purview.
This is fair enough, in view of the circumstances that quite minor errors or falsehoods in statements concerning “matter” - that is to say, concerning possibilities of experience at the very high intensity-level of tactile or visceral sensation - are always liable to lead to catastrophic experiences.
But while it is the world of “matter” which provides us with our major “disenjoyments”, many of our major “enjoyments” stem from the “unreal” world of indefinitely remote conjecture, and it is certainly well worth while to try to communicate such experiences. Any such communication is bound to be inefficient; there is unlikely to be any great degree of uniformity between the conjectural significants embodied in the several interpretants of the verbal expressions employed, but differences between them are a lively source of interest, and when they turn out to be considerable, the discovery is often very funny.
Unfortunately, the unconditional value attached to “Truth” by most leaders of human thought, since the days of Plato, has led many of their followers to attach altogether unwarranted importance to the alleged “Truth” of many statements which are entirely concerned with the “unreal” as defined in this essay. It is precisely this overvaluation which has made religious dogmatism - in which I include the Marxist variety - so very dangerous.
Scientific and religious thinkers are both mistaken in their views about these statements; they cannot be “meaningless”, but neither can they be even approximately “true”, since they permit of too great a variety of interpretants, or “meanings”. Their actualisation can never be effected, since it would have to be either in the “past” or in the indefinitely remote future, so that they can neither be corroborated nor refuted.
All we can do is to combine them with our existing stock of beliefs, including those which pertain to “reality”, in the construction of conjectural “Gestalte”. If the result is harmonious, we take the view that the degree of their falsehood is tolerable; if not, we reject them. The difference between the “will to doubt” and the “will to believe” is one of degree; the scientifically-minded “doubter” tends to attach importance even to minor discordances, while the “believer” is prepared to laugh them off provided that the total effect is pleasing.
It is a deplorable mistake to attempt to apply logical treatment to verbal expressions of the “unreal”. Dogmatic assertions of their “Truth” have constantly led to fanaticism and intolerance, while their ethical condemnation on account of their “falsehood” would lead to a disastrous impoverishment of human experience. The truth and, especially, the falsehood of statements is only important when they express significants which are within the field of “reality”, and the purpose of this essay is to show how the boundary of that field can, for practical purposes, be adequately defined.
Author's Postscript:
There is one point in this essay on which I have revised my opinion. I now think it probable that the level of intensity, or possibly “tempo”, at which “inspiration” comes to an author is a very high one. He has indeed to translate its structure into that of a medium of communication, but the process is not one of amplification and condensation, but of dilution and expansion. As far as the reader is concerned, awareness begins at the low intensity-level of remote experience, and remains there until and unless reinforced by internal “resonance”.
Editor's notes:
(1) Although hardly used in this book, “resonance” is for this author a most important word sometimes functioning in his vocabulary rather like “symbols” so crucially do for others.
(2) In The Laws of The Sun (Element Books 1996 edition; first published in Japan in 1994 - more than 7,000,000 copies sold) by Ryuho Okawa (who has had more than 300 books published) it is written: “Human souls are part of Buddha (G-d) and may be regarded as constituents of Buddha's artistic self-expression. Human beings are blessed with the capacity for creation and with freedom of choice, but many of them unfortunately misuse this opportunity to waste their lives in self-indulgence. Eventually they forget that Buddha is their origin; they forget Buddha's will and devote their lives to self-centredness and the appetites of the flesh. As soon as they find themselves more attached to this world than to Heaven, they are totally lost. In those circumstances when they return to the Real World they work to build a realm identical to the Earth they know, filled with the same desires and the same contentions. This realm is what we know as Hell” (p.116). This position is interestingly similar to that advanced by J.D. Solomon, who would, however, be less confident that: “There are ten Great Guiding Spirits of Light who embody the enlightenment of the ninth dimension… to list… their present rôles and responsibilities (as of 1994): (1) El Miore/Gautama Siddhartha/Shakyamuni/El Cantare… creation of a new age and the building of a new civilization; (2) Jesus Christ/Amor… deciding the guiding policies of the Realm of Heaven; (3) Confucius/Terabim, planning the evolution of the terrestrial spirit group and interstellar relations; (4) Achemene/Manu, racial problems; (5) Orgon/Maitrayer, the refraction of Buddha's light; (6) Kaitron/Koot Hoomi/Archimedes/Newton, science and technology; (7) Theoria/Zeus, music, art and literature; (8) Samatria/Zoroaster/Mani, moral perfection; (9) Moses/Moria… dissolution of Hell and con-troller of miraculous phenomena; (10) Enlil, guidance of the Realm of Sorcery - Arabia, the Yoga Realm - India, the Sennin Realm - China, and the Sennin/Tengu Realm - Japan; Gods of vengeance” (pp.92-3), etc.
In his recent book Reflections on Language Professor Chomsky summarises his most recent views on the nature of the faculties which have made it possible for all normally endowed human beings to make effective use of language, despite the fact that it embodies certain structural elements which are not specifically taught, but which nevertheless are successfully learned. His main contentions, I think, can be summarised as follows:-
The present writer is wholly in agreement with (i), but considers (ii) to be seriously mistaken, inasmuch as the processes which lead to the development of grammatical systems are not generative, but eliminative. This view necessitates a radical reappraisal of the opinions expressed in (iii) and (iv).
We can begin at the grass-roots by initially regarding language as a specific, highly specialised mode of vocal behaviour. This, in its earliest stages, is certainly initiated by purely physiological stimuli; the motives for our subsequent modifications of it are likely to remain obscure, but probably manifest a tendency towards experimentation, which characterises all sentient beings, and perhaps expresses an urge to achieve a greater wealth of experience, by increasing its variety. We can very well regard such experiments as examples of the “mutations” which form the actual “stuff” of evolutionary progress in the biological world. They cannot be explained in terms of anything that is already familiar, since they constitute genuine “novelty”!*
______________________
* Editor's note: Readers who have followed up my earlier references to the work, e.g., of Zecharia Sitchin and of Gilbert Ryle will appreciate that abundant evidence is available to suggest that Solomon's valuable and highly relevant account throughout this Chapter is in more than one way defective, although not as wide of the mark as is Chomsky's theory. Joan D'Arcy Cooper's account of the “guardian angel's” rôle in human growth and development also deserves to be taken into account, as does Emmanuel Levinas's key philosophical insight that the Other comes before the self (cf Robert Eaglestone, “Philosopher to the Pope” in The Tablet, 6/13 April 1996, p.472).
Of the particular group of experiments from which language has emerged, only a small proportion have survived the æsthetically-motivated processes of “human-natural” (i.e., intentional) selection, and been adopted for habitual replication in the form of “languages”. We have tended, human-naturally, to replicate the forms which have pleased us and, above all, to avoid the replication of forms which have displeased us.
Any attempt to arrive at a universal theory of language-learning by studying the forms that have survived runs into the same difficulties as that of devising a universally acceptable ethical theory by studying the mores which have survived in various parts of the world. Shaw's refutation of the “Golden Rule” pinpoints the difficulty:- “Do not do unto others as you would have them do unto you”, he wrote, “Tastes differ!”
On the other hand, the converse principle: “Do not do unto others as you would NOT have them do unto you”, is not open to similar refutation, since the basic distastes common to mankind appear to be very much alike. For analogous reasons, while it may be impossible to formulate a general theory which explains the genesis of all the language-forms which have survived, there is a good chance of discovering the general basis of the reasons which have led us, and still lead us, to reject many of the others. If we can make valid generalisations concerning the types of experience that are almost unanimously disliked by mankind, and can distinguish the kinds of linguistic experience which exemplify them today - and have presumably done so in the past - we may be able to arrive at a reasonable theory of “Universal Ungrammar”. The object of this paper is to adumbrate just such a “Darwinian” type of theory, and to apply its principles to some of the language-forms discussed by Chomsky in his book.
We can fairly assume that any sequences of phonemes which are found difficult to utter, or are æsthetically displeasing on account of their harshness or similar quality, will not appear among those habitually replicated. Such motives for rejection, however, seem to have eliminated very diverse elements in different parts of the world, and its operation has been very likely controlled by hereditary factors which have led to significant differences in the vocalising equipment. These motives are, therefore, not among the “universal” ones that we are seeking. To arrive at a general theory we have in fact, to go back to the earliest stages of vocal behaviour.
There can be no serious doubt that these are exclusively concerned with the expression of mood. The first utterance of all human beings is the wail of the newly-born infant, and this, though quite involuntary, is universally recognised as an expression of a mood of disenjoyment. Everybody who hears it will immediately interpret it as a sign of something to which his own mood would be what we can call “evitative”, i.e., the opposite of “optative”. The vocal expression of enjoyment does not begin until a good deal later; smiles are not initially accompanied by sounds but later on, soft crooning sounds become their habitual concomitants; these initially are presumably as involuntary as the wail of the newly-born.
The distinctiveness of each of these two types of sound soon becomes associated by the automatic action of memory with the distinctiveness of their accompanying moods. Once an infant has learned, probably by experiment, to control and differentiate its muscular efforts, associating a distinctive type of muscular effort with the distinctiveness of the sound that immediately succeeds it, it may well begin to replicate the sounds which have accompanied enjoyment, since they will obviously stimulate and reinforce memories of the appropriate mood. Such activities are common even in maturity; similar motives can be ascribed to an Italian peasant singing in the fields, and even to Chomsky himself, haranguing the armed soliders who had come to break up his protest meeting (op. cit., pp. 61-62) without any intention of communicating with them. We can only accept his assertion that he “meant what he said” in the sense that he felt that he had adequately expressed the strongly “evitative” mood evoked by the reports from Vietnam; it is most unlikely that he weighed his words or analysed his utterance with sufficient care to be able to justify his assertion in any very precise sense.
It is quite obvious that every child's early contacts with adult speech are interpreted, as they are intended to be interpreted, as signals of the mood of the adult. They take the form of requests, warnings, commands and prohibitions; expressions of approval and of endearment also occur, but in less urgent situations. These are less forcefully expressed and, therefore, less memorable.
All of these are intended to condition the child into associating optative or evitative moods, as appropriate, with certain distinctive characteristics of its current activity. All verbs encountered during the process will be the bare root-stems which constitute the “imperative”, for example: “Stop!”, “Come here!”, “Be quiet”, etc. “No” and “Don't” become familiar at an early stage, whereas “Yes” may not occur frequently until much later, since there is no urgency for the use of a permissive word. It functions as an optative word, but the intensity of the mood is less than that expressed by “Do!” and “Don't”; the “imperative” is our term for the highest intensities of the “optative” mood.
It would, however, be flying in the face of the evidence to regard fear as a child's sole motive for adopting an evitative attitude towards parental disapproval; sympathy has a great deal to do with it. It is obvious that unless adult sympathy for children were pretty general, the human race would stand little chance of survival, in view of the long period of maturation that children require before they can fend for themselves. The operation of sympathy is not unidirectional; most children dislike angering their parents, and enjoy pleasing them. The operation of sympathy plays a vital part in making it possible for children to progress from the mood- signalling stage of language-learning to the descriptive one.
This step is achieved by the method of question and answer. We can assume that children, on the whole, wish to please their parents, so that their choice between an optative and an evitative view of any possible activity in a particular set of circumstances will be influenced by their memory of whether such an activity was accompanied by approving or disapproving signals when the circumstances closely resembled those which currently obtain.
Here they are bound to make fairly frequent mistakes. The resemblance between the two sets of circumstances may be “superficial”; they may differ in respects which are regarded as important by the adults but have not been noticed by the child. Thus an action which may have been intended to please is followed by a disapproving mood-signal; instead of a value-experience following a crescendo of optative mood, it is suddenly replaced by a disvalue-experience. Such sudden changes from the optative to the evitative mood are what we call “shocks”. Abrupt discrepancies of this kind, between expectation and actuality, are perhaps the most acutely disagreeable kinds of experience common to all humanity. Such error-experiences give rise to doubt - the interrogative mood, a rapid fluctuation between the evitative and optative moods, which can be extremely unpleasant if unduly prolonged. Its most objectionable manifestations are what Kierkegaard characterised as “Angst”. The intensity of the interrogative mood will depend on that of the shock which has occasioned it. If this is at all severe, the child will usually exhibit signs of distress, and sympathetic adults are likely to take steps to try to prevent its recurrence.
Thus it comes about that adults, besides expressing a mood of disapproval, adopt the habit of uttering, rather prominently, a “word-sound”; this is a member of a gap-indifferent series of such sounds socially approved as an appropriate utterance in the presence of a member of some distinctive and gap-indifferent mode of experience, generally outside language altogether. The word-sound chosen will be that which is adjudged to be of distinctive importance on the occasion of the actual utterance.
Once children have grasped the importance of the association of such word-sounds with moods of disapproval and/or approval, they can call upon adults to help them, in any doubtful situation, to dispel their interrogative mood. The usual technique is to imitate the word-sound in an interrogative tone of voice, accompanying it with an ostensive gesture directed towards some prominently sensible element, usually visual, in the environment. If the word is socially inappropriate, the adult will usually answer “no!”, a sound which the child has long ago learned to associate with the evitative mood. The adult may continue by uttering, in an optative tone of voice, the word which is socially appropriate to the situation. When the child's word is appropriate, the answer will be “yes”, which expresses some degree or other of the optative mood.
It is in this fasion that children learn, piecemeal, the uncategorised lexical elements of their vocabulary. This has to be taught, not learned; there is no discernible structural correspondence between the “building-blocks” of language and their experiential counterparts, with the one exception of the onomatopœic words attached to distinctive types of sound. It is true that students of Vedantic Sanskrit claim that the phonic components of that language have a natural affinity with the moods to which their utterance is a “natural” reaction, e.g., “Ah!” to fullness, “G” to emptiness, “N” to negation, and so on; it is possible that some such connections influenced the original choice of phonemes, but few traces of them remain. The only kind of association that a child would reject would be the use of a word such as “bang” as the ostensive definition of a whimper.
There is one essential feature of ostensive definition which is very important when the words are later used for sentence-construction. The word which is being taught must be attached to an element which is currently of outstanding importance, and the word-sound itself must be highly stressed, relative to any which precede and succeed it. This is especially important in the case of relation-words, where the corresponding experiential element is a member of a gap-indifferent series of events, not an invariant “state”. There seems to be no great difficulty in diagnosing the distinctiveness of a “syntactical” transition within experience to which a relation-word is attached, provided that the necessary attention is paid to synchronising the stressed word with the transition. The outcome of the process, however, is the presence of syntactical elements among the uncategorised lexical components, and their syntactical relations to the more invariant elements involved in the ostensive definition are also remembered, just like those between the notes of a melody.
Now the syntactical relations between the more enduring elements within experience often exert an important influence on the utterer's optative/evitative attitude to them and, thanks to the operation of human sympathy, he will often desire to communicate them to his hearers in the shape of information which invariably consists of answers to questions which either have been asked or which he considers may very well be asked at a later date.
Here he is liable to run into difficulties. Speech has a syntax of its own, but its structure is extemely impoverished, since it consists of a leisurely, linear homophonic succession, and its “musical”-type resources, in the shape of variations of intensity and tempo, are very limited. They come nowhere near the possibility of exhibiting anything like the complexities of syntactical structure which utterers would often like to communicate. As long as language remains our principal medium of communication, we have to put up with this; it is the penalty for using a medium, which is basically only adapted for signalling broad, vague changes of mood, for the purpose of communicating quite minute variations in the environment which have accompanied or may accompany such changes. The only resource available is to multiply the permutations and combinations of the basic building-blocks - mostly monosyllables or disyllables - among which are included quite a variety of relation-words; these perform syntactical functions independently of their arrangement in the word-sequences. Now although we cannot hope to exhibit the syntax of our experiential subject-matter by means of the syntax of our word-sequence, there are limits to the discrepancy between them which we find tolerable. Thus, when we construct a word sequence of which the syntax differs very greatly from that which we remember as characteristic of the experiential elements associated with the conjoined words, the descrepancy is immediately noticed and produces an undesirable “shock”, which may lead to our rejection of the surface-structure of the expression as “ungrammatical” or, in extreme cases, “logically false” or “meaningless”.
Furthermore, a surface structure which permits of more than one interpretation by failing clearly to indicate the relative closeness of association expressed by that of the words and sometimes also the relative importance of their experiential equivalents may be unsuitable for resolving an unwelcome interrogative mood, on account of its ambiguity. This is normally only objectionable in cases when the ambiguity fails to resolve an oscillation between optative/evitative moods in respect of some possible action. When this is not the case, an ambiguity can be highly enjoyable; most verbal humour depends on some kind of ambiguity, syntactic or semantic, and poets frequently made use of it in order to sketch in a “harmonic” background as an accompaniment to the more obvious meaning of their “melodic” sequences. Joyce's fantasia Finnegan's Wake appears to be an attempt to express a quasi-continuous, extended polyphony in a linear language medium, making use of ambiguities which require the readers' familiarity with several languages for its proper appreciation. Of course, the “shock” occasioned by unexpected linguistic discrepancies is not intrinsically serious, but it can be sufficiently strong to give rise to physiological concomitants which can be registered “psycho-galvanically” by machines of the lie-detector type. Such machines easily register the effort we extert in suppressing our tendency to the spontaneous expression of what is uppermost in our minds, and instead amplifying some experiential memory-element which is discordant with it, maintaining an oscillation between affirmative and negative moods for long enough to utter a deliberate lie. The utterer himself is always the first hearer of his own utterance, since the feedback is almost immediate, and such “discords” are, in general, sufficiently unpleasant to evoke an evitative attitude to their replication.
At this point, the writer can summarise the extent of his agreement and disagreement with Professor Chomsky's theses, as stated at the beginning of this paper, as follows:-
The remainder of this paper will be devoted to showing how the unacceptability of the examples of “ungrammatical” language quoted by Chomsky in his book can easily be accounted for on the basis of the language-learning theory formulated above.
It is, however, necessary to emphasise that what Chomsky is dealing with is not really “language” at all, but print. This is no more than a collection of mnemonics, facilitating approximate replications of phonetic sequences when scanned in the appropriate direction. At best, it is only reported speech, and such indications as it provides of the “musical” elements of speech - the variations of tempo, intensity and pitch - are inadequate and often misleading.
For example, “It isn't going to rain” and “It isn't going to rain?” have very different spoken equivalents and meanings, but no grammatical “transformation” is needed to pass from one to the other, and the necessary difference is indicated in speech well before the end of the sentence, which is where the question-mark is printed. No visual symbolism can adequately indicate the changes of stress and intonation involved in the expression of the interrogative mood; these appear to be innate, automatic reactions, familiar to all of us. Even when a dog is the utterer, we have little hesitation in interpreting this type of intonation as the expression of an interrogative mood.
Furthermore - and this is where human beings outshine dogs - the meaning of question such as “What is that?”, which is to say, the kind of answer required, will vary according to the stress. “What is that?” requires a fuller account of a name or a description. “What is that?” expresses the high intensity of the speaker's interrogative mood. “What is THAT?” will normally be accompanied by an ostensive gesture, and requires a name or description appropriate to the visible element indicated by the gesture.Such distinctions are seldom indicated in print. Chomsky does not even use italics in his linguistic analyses in order to indicte them; his use of italics is akin to the use of case-ending in inflected languages to indicate pairs of words between which anaphoric relations may, or may not, obtain.
Now the syntactical relations between elements of direct experience include variations of quantitative intensity; especially important for language are those relating an event to its memory-echo. Any misrepresentation by the surface-structure of language of the temporal relation between the two will therefore result in an unacceptable expression. In what follows, the unstressed relation-words will be regarded as “echoes” of stressed “question” or “ostension” words; these are the forms in which the phonetic structures were almost certainly learned. Similarly, the unstressed copula will be regarded as an echo of the assertive is!, which may in turn perhaps be regarded as an echo of the interrogative, highly stressed IS?
The first of Chomsky's examples to be examined in accordance with the above principles will be one that appears on p.86 of his book:
Chomsky writes:- “There is good evidence that (6) - presumably, its meaning - derives from the underlying structure (6'). (6') Y seems [s John to be a nice fellow].”
We are immediately confronted with a paradoxical situation; whereas (6) seems quite natural, the expression which is supposed to correspond more closely with its meaning is quite unacceptable. The reason for the discrepancy is quite simple. (6) expresses direct thought, whereas meaning, for the hearer, is “reported thought”. The situation is made clear by the use of the “accusative-infinitive”, the construction used in Latin for the expression of reported speech, and the variable “Y” stands for whoever is the speaker. Substitute “I” for “Y” and we obtain:- (6") I seem(s) [s John to be a nice fellow].
Here, however, the agent-patient relationship feels quite wrong. If we invert it, writing [s John to be a nice fellow] seems to me, it is still unsatisfactory, becuase it is I who judge him to be a “nice fellow”.
A sequence of events is actually involved in the first of which I am “patient”, in the second, “agent”. It can be described as follows:-
(a) An extremely complex compresence of sense data and memories “shows itself” as “subject” to me as “object”; it includes a memory of the word-sound “John”, since I remember being taught to take an optative view towards assocating a member of that distinctive gap-indifferent series of word-sounds with any member of a certain distinctive gap-indifferent series of complex structural “Gestalte”, and such a member forms part of my present awareness.
(b) It is usually important, under such circumstances, to decide whether to take an optative or an evitative view of such actions as may lead to the replication of experiences similar to the present one. My present experience is pleasing, so I seek to communicate my mood to my hearers by associating the words “nice fellow”, which I have been taught to use for this purpose, with the word “John”. I proceed to utter the words in the same order as that in which the events have occurred. A summarised account could be: (i) “John” appears to me; (ii) I deem “it” a good fellow. More briefly still, omitting the personal pronouns (which can be taken for granted): “John seems to be a good fellow.”
(6") is ungrammatical because it misrepresents the nature of the agent-patient relation, an element in the “syntactical structure of experience” with which we may fairly claim to be directly acquainted.
“John seems to Y to be a good fellow” is from the hearer's point of view less ungrammatical than (6"), but still fails to indicate Y's rôle as agent of the judgment “to be a good fellow”. It has the merit of indicating the correct order of the events involved.
It is possible to treat “seems” as an “apponent” verb, an active form fulfilling a partly passive function; “is deemed” would appear to be an adequate substitute in the passive voice, so that (6) can then be written:- “John is deemed to be a good fellow.”
We can now, if we like, reintroduce the variable Y as follows:- “John is deemed by Y to be a good fellow.” This seems to be altogether unexceptionable, and since “me” can stand as a value of the variable Y, the form is satisfactory both for direct and reported speech.
The semantic function of “to be” calls for some comment. Superficially, “John seems a nice fellow” does not appear to differ semantically from “John seems to be a nice fellow”; but “to” in English always has a connotation of purpose - of future possibilities. Its use here suggests that the present instance of “deeming” is likely to be replicated. When it is omitted, it becomes clear that the simpler form “John seems a nice fellow” is a member of the following transitively ordered series:-
A normal intensity of the affirmative mood, characteristic of the “deadpan” level of written “language”, is expressed by (b). In (a) the importance of the affirmative mood temporarily exceeds that of what is affirmed; in (c) it is weaker, in (d) weaker still. The stress placed on SEEMS serves to express the greater departure from the “norm”; we should expect the next sentence to begin, “but, only the other day, he…!”
It is notable that whenever the strength of the affirmative mood attains or exceeds the “deadpan” level, the persistence expressed by “to be” in association with “seems” is taken for granted. Its habitual omission has led to the association of “is” with “be”, which is probably a complete mistake. We have seen that “seems” can be advantageously replaced by “is deemed”; on similar grounds there is a good case for replacing “is” by “is affirmed” in order to clarify its semantic function. It then appears to be a purely linguistic “performative” quantitising an affirmative mood on the utterer's part towards a compresence of semantic units that it currently conjoins.
In its “imperative” form (stressed, at the beginning of a sentence) it calls upon the hearer to affirm the word-sequence that follows, pronouncing that sequence with the intonation characteristic of the interrogative mood. The hearer may simply reply “yes” or “no”; otherwise he will repeat the word sequence, inserting one of their synonyms, “is” or “is not”, slightly stressed, at the point where there has been a perceptible pause in the speech. A child who has learned to use “is” has no difficulty in understanding the structure of a sentence such as “The man who is tall is in the room”, because the first “is” (a remote echo of the former affirmation) is slurred over rapidly, whereas the second, preceded by a slight but discernible pause, is clearly the present “performative”. The phrasing is effected by “musical” means, which do not appear at all in print!
The word “be” is an altogether different matter. It is derived from the Celtic “byw” and closely related to the Greek “bios” (i.e., “life”). Used in the imperative mood, the form in which a child will almost certainly learn it, it expresses a strongly optative mood towards some state of behaviour on the child's part, as, for example, in “Be quiet”, which could be paraphrased as “Live quietly!” It never calls for linguistic behaviour. The Latin “esse” is not the equivalent of “be”; as far as I am aware, it is used exclusively in reported speech and thought* and is clearly an abbreviation of “est se”, i.e., “is affirmed itself”. This can only be applicable to some kind of linguistic expression.† “Est” in Latin is not ordinarily employed as a copula but is used where emphasis or quantification is required. Its rôle in Russian is also that of a quantifier.
______________________
* Editor's note: In mediæval Latin writings about theological and metaphysical questions esse is used much more variously than Solomon here supposes.
† Editor's note: or speech act!
The sentences on pp. 86-87:
are similar in form to (6), and their analysis is:-
affirmed certain
John yes (by Y) to win.
deemed to be certain
The only ungrammatical expression discussed in this section is “John seems to be probable to win.” The trouble here is syntactico-semantical, as is clearly shown by the fact that it can be cured by substituting “likely” for “probable”. No existing state of affairs can properly be described as “probable” (which is semantically tied to the “future”), whereas “likely” expresses a degree of the speaker's present affirmative mood. This semantic distinction might be useful to the probability-theorists, who appear to have ignored it.
We will now pass on to the examples given on p.85 of Chomsky's book, viz.:-
The semantics of the acceptable versions clearly imply that the following sequence of events has occurred:
Events (a) and (b) occurred a long time ago, and their sequence is quite unimportant; neither (2) nor (3) indicate it quite correctly. But event (c) is recent, and the surface-structure of (5) reverses the order of (b) and (c) so that it is felt to be unsatisfactory. Best of all would be the following version which Chomsky does not give:- “Of Tolstoy's novels, the only one that I like is out of print.” (5) is not intrinsically ungrammatical. It is a rather clumsy paraphrase of:- “Tolstoy's only surviving novel is not being printed in a way that I like.”
In this example the ungrammaticality is brought about by a simple inversion of the temporal sequence implicit in the semantics. In the next one, an ambiguity of mood is involved. The following appear on p. 88:-
We can once more proceed to list the sequence of events implied by these statements, beginning with the more acceptable one (11):
At this point, the police know “who” and therefore, in the sense of “remember”, can also think “who”. But before event (f) occurred, they were thinking “who?”, so that there is here an uncomfortable ambiguity between the interrogative and affirmative moods, owing to the ambiguity in the meaning of “think”.
The question “Who do the police know the FBI discovered that Bill shot?” is also uncomfortable for analogous reasons. What is “known” is not normally a “Who?”, although it may be for the actual speaker.
The explanation of the examples on p.89 depends on the circumstance that the relative, unstressed “who” always requires an implicit or explicit stressed “who?” of which it is the “echo”. As none of the sentences has more than one “who”, we can take the questions forms (13") and (14") as primary. The problem now is as follows:- If from (13"), “Who? did John discover that Bill had seen” we can derive (13'), “John discovered who Bill had seen”, why from (14"), “Who? did John discover pictures of” can we not derive: “John discovered who pictures of”?
The answer is extremely simple. (13") implies that John as asked “who? has Bill seen”, so that the relative “who” is in order. In (14") it is only the speaker who is asking Who?, and unless there are some signs that an answer has been forthcoming no relative “echo” is meaningful.
In examples (15) and (16) on p.90, the problem to be solved is as follows:- From (15), “John believed Mary said Tom saw someone” we can derive the question (15'), “Who did John believe that Mary said that Tom saw?”, but from (16), “John believed the claim Tom saw someone” we cannot derive the question (16'), “Who did John believe the claim that Tom saw?”
Chomsky in accordance with his “trace” theory writes “who-someone”, but the “who” is in the wrong place; this is only in order if there has been no mention or enquiry of a name, and this is certainly now what Tom saw [cf examples (9") and (11) above]. Mary may have mentioned a name, in which case the “who” is in order in (15) although it should come after “Mary said”, not after “Tom saw”. But the “claim” was actually “Tom saw someone”; there is no possibility of introducing a name here. (16) could be written: “John believed the ‘Tom saw someone’ claim”.
The difference is between the temporal structures implied by the semantics; (15) expresses the event-succession “Tom saw” and “Mary said”, whereas (16) expresses an enduring compresence of “the claim” and “Tom saw someone”.
We can next consider the active-passive pair on p.97:-
In the case of (20) there is an unescapable temporal relationship between the experiential equivalents of the words; the compresence of
beavers precedes the presence of dams.
building
Later, a suitable dam-experience will evoke a memory of
building
beavers - the thought succession being now expressed by
"Dams, yes, beavers".
built
We have already given reasons for rejecting the supposition that “are” expresses any more than a current affirmation. “Yes” effects the required conjunction without the ambiguous association with “being” which suggests its perpetuation.
To effect the spoken expression of the thought, we have to choose a sequence for the bracketed words. “Dams are beaver-built” is not incorrect, but “built” is more closely, i.e., more frequently associated with “dams” than is “beavers”, so we prefer the order: “Dams, built, beavers”. This, however, is impermissible as it stands, because it inverts the basic temporal relationship between the experiential equivalents of the uncategorised lexical components. The difficulty has been overcome, at some time in the past, by concocting an adverbial phrase, which is equivalent to “beaverly” or “beaverlike”. The prepositional prefix “by” is used for the purpose, and is written separately from “beavers” although in speech “by beavers” is virtually a single word, equivalent to the Latin “castoribus” (which employs a suffix for the same purpose). This “adverb” is not related to the other words by any specific temporal relation, so that the syntactical objection is removed.
We can now see why the question “What are dams built by” occasions a certain feeling of discomfort; the “ibus” has been detached from the implied “castor”. Preferable is “By what are dams built?” “By what?” is a sub-species of “How?”; the “what” is stressed, but not the “by”. Since, however, the syntactical discrepancy is intra-linguistic, it can be dismissed as humorous rather than serious. If the “incorrect” version is rhythmically preferable for the purposes of speech, it should be adopted. Churchill's classic “This is the kind of language up with which I will not put!” is a superb expression of this point of view, by a master of the spoken word.
The “split infinitive” is open to the analogous objection that the adverb weakens the frequently purposive nisus between “to” and the verb. The degree of weakening increases with the actual length of the adverb; “to truly believe” is harmless, but “to incon-sequentially believe” is awkward.
Example (23) on p.99 is an excellent example of the difficulty arising out of the circumstance that whenever an utterance embodies an egocentric particular, its meaning for the hearer is not the same as for the utterer. Let us suppose that the answer to (23), Who? said Mary kissed him, is “John”; he very likely said “Mary kissed me.” But the answer to Who? must be a “he” since in this case “I” and “you” are ruled out by “Unlike-person constraint”, which certainly appears to be a rule of Universal Ungrammar. Provided that “him” is unstressed relative to “Mary” and “kissed”, then “who?” and “him” will be generally interpreted as co-referent, simply because “John said Mary kissed him” is the only way of reporting John's utterance. If, however, “him” is at all highly stressed, acquaintance with the total speech-situation will be needed to remove possible ambiguity.
In the case of (24), Who? did he say Mary kissed, the answer “himself” is the only one, out of an indefinitely large selection of possible “he”'s, which will establish co-reference between “who?” and “he”. This is unlikely, though by no means impossible.
P.101:
“Each other” is an adverbial phrase, a conveniently brief way of expressing a plurality of quasi-simultaneous relations. In most contexts, it is replaceable by “one another” or by “reciprocally”. it always requires a plural subject and verb-form, and fulfills no semantic function without them.
Once the phrase is split into its constituent parts, the quasi-simultaneity of the relations is disrupted, so that singular subjects and verb-forms are in order. The only grammatical example of the adverbial usage given by Chomsky does, in fact, provide a plural subject: (33), We seem to like each other. The only “abstract structure” that we need to appreciate in order to reject (29) and accept (29') is the difference between a quasi-simultaneity and a succession. It is difficult to imagine even the most dyed-in-the-wood empiricist denying that such “knowledge” must be “innate”: but if we are to characterise such structural differences as “abstract”, whatever are we to regard as concrete?
P.107:
There is no need whatsoever to derive this from the “factual” statement (43), A book is on the table. All we need to remember is that “There!” is an echo of the ostensive “THERE!”, and “is” of the affirmative “yes”. The thought-sequence: “THERE! yes, a book on the table” is perfectly in order. The answer to the question “WHERE? a book” is “a book THERE yes on the table”; since the whole judgment is of a temporal compresence, almost any sequence will do - provided that “There” and “yes” are contiguous. Location is mentioned first whenever it is likely to be the most useful part of the information conveyed. “Is” must, however, be kept away from the beginning of the sentence, since this would express the imperative “affirm!” The question equivalent to “A book is on the table” is “Is a book on the table?” This can be answered directly if the table is within the hearer's field of vision, but we are so accustomed to speaking of conjectured or remembered tables, in which case “there” expresses “other than here”, that we habitually insert the word even when it is redundant.
Examples could be multiplied indefinitely, but the above should suffice to show that the theory of “eliminative ungrammar” is sufficient to account for the unacceptability of certain language forms; the complex rules of Chomsky's “generative grammar” can very likely be subsumed as special cases in the more general theory.
The theory of eliminative ungrammar certainly presupposes an innate appreciation of structure; it is, however, highly questionable whether the structure concerned should be thought of as “abstract”. It is, broadly speaking, of the type characteristic of music, including temporal succession, rhythm and changes of intensive quantity. Now structural contrasts, degrees of difference, relative tempi and intensities of change are the very stuff of all experience. But it is invariances and gap-indifferences which furnish the stock-in-trade - the uncategorised lexical components - of language, because it is these to which the gap-indifference of word-sounds and ultimately the invariance of written symbols can be attached.
For language, therefore, it is the structures of change which appear “abstract”, whereas for direct experience it is the relatively invariant elements which are abstracted for the purpose of symbolisation. Do we regrd the picture on the television screen as concrete? If so, we probably tend to regard the transmission-wave as “abstract”, even though we believe that the structures visible on the screen were embodied in those of the carrier-wave from which they have been “abstracted”
It is questionable whether the formulation of generative “rules” for grammar is at all desirable; such rules, once formulated, may tend to impoverish language by restricting the range of experimentation. The few simple principles underlying “ungrammar” would appear to be adequate for remedying or eliminating unacceptable language-forms. The notion that rules are intrinsically good things is a superstitious belief, unfortunately widespread among the more “orthodox” thinkers, both scientific and religious. Einstein's “The Old Man doesn't play dice” is an eloquent expression of this belief. It has had far too long an innings; a much more joyful point of view is that “He” plays dice the whole time, but, being a sentient “Being”, tends only to replicate the throws that “He” likes and avoids replicating those that “He” dislikes. In effect, this means that such “determination” as exists in the Universe is of æsthetic rather than mechanical origin.
Such is the view that I have taken of the development of the linguistic faculty in human beings. Granted our inherent propensity for vocal experimentation, the rest can be explained by æsthetically motivated selection which, in so far as it is guided by basic dislikes, is indeed “universal” for humanity.
- Shalom & Welcome! - 
© The Neith Network Library 2002
Webmaster: H.B. ExtraReverendDoctorColinJames Hamer, The Rainbow Programme
Creativity House, 9 Oxford Street, St. Thomas, EXETER, Devon EX2 9AG, U.K.
This page updated 00:01 1/8/2002.