- In my “Criteria for Style Analysis” (hereinafter Criteria), Word XV (1959), 154–174, esp. §§3.0–3.2.4; see also my review of S. Ullmann's Style in the French Novel, in Word XV (1959), 404–413, esp. pp. 407–409.
- On this variability of effects, or polyvalence, see my review cited footnote 1, page 409. Similarly, L. Spitzer, after comparing—with a very different approach—more styles than any stylistician ever had, concluded that the stylistic sign is “empty” (see, for example, Modern Language Quarterly, XIX [1958], 235, footnote 8).
- See Criteria, §§1.1–1.2.
- The size limits of this type are very narrow: a poem hardly longer than a monostich, such as the Latin elegiac distich, would not fall in the same category, since the complexity of the metric pattern makes up for its brevity. Quotations from other authors must also be excluded, because they are provided with a new context. Dictionaries of quotations, on the contrary, should be included: they are meta-texts only for the style analysis of a literary work as a whole, but isolated, without consideration of the compiler's reasons for excerpting them, they may be studied to determine to what extent a context can be cut off without destruction of the stylistic effect.
- On this phenomenon and its name, see Criteria §3.2.3. It is encoded in the written text and must not be confused with the concurrence of several sets of signals, such as can be observed in the performance of certain oral literary forms (see T. A. Sebeok, “Folksong Viewed as Code and Message,” Anthropos LIV [1959], 141–153, esp. 141–42).
- On perception in reading (word recognition, peripheral vision, etc.), see the handy summary of M. A. Tinker “Visual Apprehension and Perception in Reading,” Psychological Bulletin XXVI (1929), 223–240; the yearly summaries of “Reading Investigation” by Wm. S. Gray in the Journal of Educational Research.
- It may be an almost unconscious re-reading by peripheral vision; or a conscious search for the solution of some decoding problem: this search entails several simultaneous Markov processes working forward and backward (see a good example in A. A. Hill, “An Analysis of The Windhover: An Experiment in Structural Method,” Publications of the Modern Language Association LXX (1955), 968–978, esp. 975–976).
- It is obvious that I do not start my analysis from this unmarked element (at the present stage, this would amount to falling back upon the concept of norm), but from the contrast immediately perceived by the reader; for a discussion of this approach, see Criteria, Chap. 2.
- Like this paper, the examples given are orientational: they are introduced as further clarification; but no measurement will be attempted in the present study. Once again, I am trying only to provide a coherent organization of materials, using nothing but relevant features.
- Note that clair-obseur does not appear in French until 40 years later, and then only in its Italian form and in technical parlance: there was no association susceptible of weakening the SD.
- See L. Spitzer, Aufsätze zur romanischen Syntax und Stilistik (1918), pp. 289ff.
- This example is borrowed from Justin O'Brien, “Proust's Use of Syllepsis,” PMLA LXIX (1954), 741–752; this article can be used as a monograph on syllepsis in general. On zeugma, closely related to syllepsis, see G. O. Rees, Français Moderne, XXII (1954), 287–295; R. Le Bidois, ibid., XXIV (1956), 81–89, 259–270.
- Cf. the related question of the duration of stylistic effects through successive “performances” of the text (see Criteria 1.3.0–1.3.1; Wellek and Warren, Theory of Literature, 1956, Chap. XII). The conditions of unpredictability are maintained in the “frozen” structure of the message; thus successive performers will be exposed to the same phenomenon. For a reader “performing” repeatedly, memorization will cancel unpredictability, but will preserve the formal characteristics of the text. Whether memorized or unexpected, style remains an active factor. As for literary imitation, it renews the SD by transferring it to a new context (e.g. Vergil using Homeric devices): this last process is no different from the use of a cliché in a new environment.
- A. Note that in this respect microcontext can be relatively long: even a repetition, for instance, can be the immediate constituent of a SD; the longer the repetitive pattern lasts before being broken, the stronger the rupture effect—B. Of course we must distinguish: (1) the repetition as SD, that is, a contrast-creating structure which consists of the rhythmic or arhythmic recurrence of an element (e.g. a verbal ending); (2) the repetition as microcontext; (3) the repetition of SDs, that is, the recurrence of the same stylistic structures with various contents, this being a case of loose terminology, the normative metalanguage of traditional literary criticism—C. I use here (and elsewhere) the verb prepare. I mean it to describe a context building toward a SD (e.g. in a period, free rhythms preparing the contrasting clausula, or a long protasis preparing the antithetic climax of a short apodosis); I never mean it as a statement of the writer's intentions.
- Continuity should not be taken as an assumption about the author's psychological processes. Without doubt the pattern corresponds to a certain continuity in the writer's mind (true even of écriture automatique in French Surrealistic writings: this device was in deliberate opposition to the continuity produced by reflection, but it aimed nevertheless at recording the continuity of subconscious mental activity), but it does not depict it in the form of a sequence of words; to believe so would be a metalinguistic confusion (cf. the “chronological” interpretation of Latin word-order in sentences like urbem captam hostes diripuerunt).
- See Joshua Whatmough, Poetic, Scientific and other Forms of Discourse (1956), pp. 105–107.
- We find many similar instances in the interplay of narrative/direct discourse/indirect discourse/ free indirect discourse; see S. Ullmann, Style in the French Novel, pp. 94–120; bibliography of this device, in A. G. Landry, Represented Discourse in… Mauriac (1953), pp. 76–79.
- It could be said that this type of context is close to the linguistic norm. The departure, however, is not from the “consciousness of the norm,” but from a written background.
- “Editors classify these deletions indiscriminately under the head of concision; but this involves the introduction of the idea of simplicity, which is an esthetic criterion hard to define and irrelevant in an inquiry as to the existence of style; the deletion amounts to an increase, objectively measurable, in the distance between SDs. E.g. Balzac, who is far from aiming at simplicity, deletes many emphatic devices (tout modifying adjectives, c'est… qui/que…, etc.). These features, frequent in speech, are dropped significantly from written conversations: without intonation, they would only be padding, and their omission permits a rapid succession of sudden contrasts more suggestive of the vividness of actual speech than a total recording would be. See Mario Roques, “Manuscrit et éditions du Père Goriot,” Revue universitaire XIV (1905), 34–42, 71–76, 178–183.
- Needless to say, this is a very partial analysis. It does not take into account rhyme and rhythm, or concurrent and overlapping SDs such as (context) nor…/only…, which in turn contains nor any,../nor….
- If the devices were identical in form and content, we would have one SD: repetition (cf. footnote 14). The reasons for these high frequencies range from unconscious or conscious self-imitation (on stylistic contagion, see L. Spitzer, Linguistics and Literary History, p. 191, n. 46; J. Marouzeau, Revue des études latines XIV (1936), 58–64; XXVI (1948), 105–108) to a reaction against normative precepts forbidding repetitions. We shall have to inquire whether there are SDs whose microcontext makes them immune to saturation (e.g. disjunctive word-order, because it suspends decoding altogether) and we shall need statistics, based exclusively on stylistically relevant data. This is in synchrony a problem similar to that of the duration of SDs in diachrony (i.e. resistance to wearing out through repeated performances and “traditionalization”; see, for example, M. Riffaterre, “La durée de la valeur stylistique du néologisme,” Romanic Review XLIV (1953), 282–9).
- The mot juste, for example, may be defined by its fitness, whose measure in language is statistical (J. Whatmough, Poetic… Discourse, pp. 119–121). But its stylistic relevance does not lie in its fitness, which is defined in relation to the linguistic structure; it lies in the variability of its effects in relation to different contexts (and to different poetic dogmas). The mot juste may be a SD, or it may create a context so “fitting” that vague, “unfitting” words will become devices by contrast.
- For example, in Latin, see Bertil Axelson, Unpoetische Wörter, Lund, 1945.
- It is significant that poetic language frequently uses Umgangssprache forms: they would have no effect in their usual context; they have an effect in the contrived setting of meter. Pointing in this direction are observations such as Meillet's that “la poésie (latine) n'a pas le purisme de la prose” (Esquisse d'une histoire de la langue latine, 1938, p. 220); prose, unprotected by conventional forms, was not able to use “ordinary” words without losing its literary character. (Cf. frequent remarks in poetic glossaries that certain words belong equally to the “poetic and familiar” vocabularies: clearly they are SDs in the former, contextual in the latter). See in French the use of substandard speech forms l', m', j', n' (le, me, je, ne) in Jules Laforgue's poems. Normative conditioning of the literary message, which linguistic analysis bypasses, should become apparent (see L. Kukenheim “Réflexions non-structuralistes,” Neophilologus, 1955, 161 ff).
- See Criteria, p. 169, n. 24 and footnote 12. (These must not be confused with situational norms, predetermined by the topic.) For example: elements suggesting (not transcribing) either foreign pronunciations (e.g. speech of German characters in Balzac, of Jews on the English Victorian stage), or idiolectal features used for psychological characterization (e.g. Mr. Jingle's way of speaking in the Pickwick Papers), or social attitudes (e.g. parodic use of stereotypes; see M. Riffaterre “Sur un singulier d'André Gide. Contribution à l'étude des clichés,” Français Moderne XXIII [1955], 39ff.), or other authors' styles (e.g. Proust's pastiches; imitations of the Bible; for the latter even one feature may suffice, see L. Spitzer, Linguistics and Literary History, p. 150; Uriel Weinreich, “On the Cultural History of Yiddish Rime,” Essays in Jewish Life and Thought, 1959, p. 423ff., esp. 434–435; on a much more complex mimetic standard, see Y. Le Hir, Lamennais écrivain, 1948, pp. 241–332). A good example of SDs which create a nonce- standard is the hyphenated neologisms Carlyle employs in Sartor resartus to suggest “German-philosophical” style.
- Choose new content alerts to be informed about new research of interest to you
- Easy remote access to your institution's subscriptions on any device, from any location
- Save your searches and schedule alerts to send you new results
- Export your search results into a .csv file to support your research