Cotextual Cues for the Annotation of Perspective in Livy’s Ab Urbe condita

(10) Quibus caesis omnibus praeterquam hostili odio etiam ne quis nuntius refugeret, extemplo Scipio neglectas ut in tali tumultu portas inuadit; ignibusque in proxima tecta coniectis effuse flamma primo uelut sparsa pluribus locis reluxit, dein per continua serpens uno repente omnia incendio hausit. (Liv. 30. 6. 4-5)

“When they had all been slain, not only because of an enemy's hatred, but also that no man might escape to tell the tale, Scipio at once burst into the gates, naturally unguarded in such a commotion. And then as firebrands were thrown upon the nearest roofs, the flames pouring out at blazed as if scattered over a number of locations; and then creeping along without a break they promptly consumed everything in one conflagration.”

(11) Caesa aut hausta flammis ad quadraginta milia hominum sunt, capta supra quinque milia, multi Carthaginiensium nobiles, undecim senatores; signa militaria centum septuaginta quat- tuor, equi Numidici supra duo milia septingentos; elephanti sex capti, octo ferro flammaque absumpti. (Liv. 30. 6. 8-9)

“Slain or burned to death were some forty thousand men, more than five thousand captured, many Carthaginian nobles, eleven senators. Of military standards a hundred and seventy-four were taken, of Numidian horses over two thousand seven hundred. Six elephants were captured, eight destroyed by sword or by fire.”

In example (9), the three subordinate clauses indicate the ability of an observer -- with an overview of the events taking place below their “top-of-the-hill” position -- to make comparisons to other times and places (P5; see section 2). The same perspective was annotated for example (10), where phrases like ut in tali tumultu “as (natural) in such a commotion” are considered externally evaluative). In example (11), the evaluation of the casualties of the battle is by definition dependent on a P1, concluding the episode and looking back on the events with hindsight.

The correlation plots in Figure 11 and Figure 12 show the standardized residuals of the X² Tests of Independence that evaluated the correlations between the annotation of perspective and COMP (N = 595; X²=92.97, df = NA, p < 0.001; see note 33) and EVA (N = 588; X² = 279.12, df = NA, p < 0.001; see note 33) as highly significant.

The hypothesized correlations (see section 3.1) seem to be confirmed by the data, with the exception of the one between external EVA and P5, for which the positive correlations (in grey) are shared between the absence of all EVA and the presence of external evaluation.

In addition, and in contrast to their role apparent from the first part of our analysis (but see note 30), a significant correlation was found for the variables PERS (N = 595; X.² = 21.87, df= NA, p = 0.001; see note 33) and DESC (N = 595; X2 = 102.89, df=4, p < 0.001). Especially the presence of DESC was very strongly negatively correlated to P1 (stdres = -9.18), and (to a much lesser extent) to P2 (stdres = -2.26); a significant positive correlation seems to exist between DESC in the cotext and P3 (stdres = 3.13), P4 (stdres = 5.11) and P5 (stdres = 4.96). Example (12) illustrates the positive association between P5 and the presence of descriptive and external EVA:

(12) Hannibal spem nanctus locis natis ad equestrem pugnam, [qua parte uirium inuictuserat,] See note 57.

⁵⁹ The interpretation of iam as “presently, now” in this passage is due to the tense chosen for the main verb in this clause: in Livy's narrative, iam in the sense of “already”, i. e. as an aspectual temporal adverbial (see note 50), occurs almost exclusively with the impf. ind. and the pluperfect tense in our corpus. Although the pf. ind., one of the morphologically possible interpretations for the tense of uenit, is used in temporal subordinate clauses (e.g. with postquam) with the meaning of, indeed, a pluperfect indicative already in Livy's work, the use of the pf. ind. with past-before-past meaning in main clauses is very rare before the development towards Late Latin (cf. Kiss 1982, 11). The interpretation of uenit as a pr. ind. and iam in the sense of “now” seems much less far-fetched.

⁶⁰ See also note 48.

^{facturos copiam pugnandi consules, dirigit aciem lacessitque Numidarum procursatione hostes. (Liv. 22. 44. 4)}

“Hannibal, having conceived hope that the consuls would give him the opportunity of fighting in a location naturally suited for a cavalry fight, [in which part of vires he was unbeaten,] formed up his battle line and harassed the enemy with a sally of the Numidians.”

Fig. 11. The impact of the presence (“yes”) or absence (“no”) of comparative lexis on the annotation of perspective

Fig. 12. The impact of evaluative lexis (”internal” or “external”) or its absence ('no”) on the annotation of perspective

P5 is indicated by the words spem ... consules “having ... fight”, which are considered as an evaluative remark on the part of the narrator, and the words [derigit] aciem “[formed up] his battle line” and procursatione “a sally”, which appeal to the perceptive senses of the audience.

For the presence of PERS in the cotext, a significant positive correlation was found to P2 (stdres = 3.79) and a significant negative one to P1 (stdres = -2.44). In example (13), the positive association between P2 and the presence of PERS is illustrated:

(13) Nec Scipio ullo tempore hiemis belli opera remiserat, quae multa simul undique eum cir- cumstabant. Vticam obsidebat; castra in conspectu Hasdrubalis erant; Carthaginienses deduxerant naues; classem paratam instructamque ad commeatus intercipiendos habebant. (Liv. 30. 3. 3-4)

“Nor had Scipio, at any time during the winter, lessened his military operations, which were many and all around him at the same time: he was besieging Utica, the camp of Hasdrubal was in [his] sight, the Carthaginians had launched their ships, and were keeping their fleet ready and equipped, in order to intercept [his] supplies.”

The perspective of Scipio, whose name is placed (very conspicuously) at the outset of this new passage and who is thereafter constantly referred to with pronouns (“pronomi- nalization” or “subject-chaining”, cf. Stockwell 2002, 53-54), is apparent by instances of PERS such as opera ... quae eum circumstabant “many [concerns] were around him” and in conspectu “in [his] sight”.

In the fifth and sixth tests of this second part of our analysis, the hypothesized correlations were evaluated between the annotation of perspective on the one hand, and deictic indications (illustrated for P5 in examples (6) and (14)) and for P1 in example (7)) and deictic tense usage (illustrated for P1 in example (7)) on the other hand (see section 3.2).

(14) Multi mortales conuenere, studio etiam uidendae nouae urbis, maxime proximi quique, Caeninenses, Crustumini, Antemnates; iam Sabinorum omnis multitudo cum liberis ac co- niugibus uenit. Inuitati hospitaliter per domos cum situm moeniaque et frequentem tectis urbem uidissent, mirantur tam breui rem Romanam creuisse. Vbi spectaculi tempus uenit deditaeque eo mentes cum oculis erant, tum ex composito orta uis signoque dato iuuentus Romana ad rapiendas uirgines discurrit. (Liv. 1. 9. 8-10)

“Many people ... came together for the festival, especially those who lived nearest, the inhabitants of Caenina, Crustumium, and Antemnae. Now the Sabines came with all their people, including their children and wives. ... When the time came for the show, and people's thoughts and eyes were busy with it, the preconcerted attack began. At a given signal the young Romans darted this way and that, to seize and carry off the maidens.”

In example (14), the PROX conuenere “came together”, proximi “the neighbouring peoples”, iam “presently”,⁵⁹ uenit “came” (x2), discurrit “darted this way and that” strongly suggest a vantage point taken, by the narrator and their audience, at the scene of the events within the story-world.

The tests evaluating the independence between the annotation of perspective on the one hand, and deictic indications, if present at all (N = 140; X² = 125.24, df = NA, p < 0.001; see note 33), and deictic tense usage (N = 564; X² = 46.37, df= NA, p < 0.001; see note 33) on the other hand, were statistically significant. The correlation plots in Figure 13 and Figure 14 represent the standardized residuals of these tests.

Based on the relevant samples, the presence of PROX appears to be significantly positively correlated to the annotation of both P4 and P5, while the same seems to hold true for DIST and P1. In addition, deictically used tense forms are very likely to occur especially in the cotext of SoAs narrated from a P1, but quite unlikely to occur if the perspective is P3, P4 or P5.⁶⁰

The last test of the third part of our analysis, concerning the hypothesized negative correlation between P2 and the presence of a temporal adverbial (see section 3.1), indicated that this correlation is not statistically significant within our current corpus (N = 584; X.² = 18.06, df= NA, p = 0.32; see note 33).

Fig. 13. The impact of deictic indications (“proximal” and “distal”) on the annotation of perspective

Fig. 14. The impact of the absence (“no”) or presence (“yes”) of deictic tense usage in the cotext on the annotation of perspective

Concluding remarks

The aim of this paper was to ascertain whether the proposed methodology of “cotextual cues” for the annotation of perspective is confirmed by the language data I have gathered so far. More specifically, our goal has been to evaluate the strengths of the correlations between the cotextual cues reported in the literature (see section 3) and the observable variables of verb tense selection on the one hand, and the annotation of perspective on the other. Performing statistical tests (the Conditional Inference Tree and the X² Test of Independence) on the relevant samples, we have shown that the correlations between verb tense selection and lexical cues -- apart from personal and descriptive lexis, and only for the pf. ind. and impf. ind. (but see note 13) -- were as we had expected them to be. However, for mental processes in particular, the corpus data did not yield the hypothesized correlation: the many relational processes in the corpus with era(n)t probably influenced the correlation between mental processes and the impf. ind. The hypothesized correlations between verb tense selection and deictic cues were also largely confirmed, especially for the pf. ind. and pr. ind. (but see note 15), as well as those with subject animacy (see note 16).

Next, the correlations between the verb tenses and perspectives were shown to be positive/negative and strong where we had hypothesized them to be -- despite the principle of metafunctional hierarchy (see note 30).

Finally, all cotextual cues were shown to be correlated to the annotation of perspective as we had anticipated in section 3, except for subject animacy (see again note 16). Concerning process types, the issue with mental processes and the impf. ind. appears resolved: mental processes are strongly correlated to the scenic-camera eye perspective, which, indeed, has many events in the historical present tense when foregroundedness is stressed (see note 30; Aerts submitted, section 3). Interestingly, verbal processes seem correlated to a viewpoint from within (P3 or P4) rather than to a temporal illusion of proximity (P4 and P5); indeed, to hear a character's words implies being cognitively closer than, for instance, on top of a hill. Similarly interestingly, material processes appear to be positively correlated only to P5.

In conclusion, apart from these minor issues, most of which are explainable within our wider three-dimensional framework (Aerts 2018; Aerts forthcoming; Aerts submitted), no significant conflicts were found in any test regarding the value and validity of our framework of cotextual cues for the annotation of perspective in Livy's narrative.

latin narratives cotextual cues verb perspective

References

Adema S. M. Tenses in Vergil's Aeneid. Narrative Style and Structure. Leiden, Brill, 2019.

Aerts S. Tense, Aspect and Aktionsart in Classical Latin: Towards a New Approach. Symbolae Osloenses 2018, 92, 107-149.

Aerts S. A Three-Dimensional, Systemic Functional Analysis of Tense Usage in Gregory of Tours' Historia Francorum. Acta Antiqua Academiae Scientiarum Hungaricae 2019, 59, forthcoming.

Aerts S. An SFL-Inspired Corpus-Based Analysis of the Combined Impact of Three-Dimensional Latin tense and Aspect on Verb Tense Selection in Livy's Ab Urbe Condita, submitted.

Allan R. J., I. J. F. de Jong, and C. C. de Jonge. From Enargeia to Immersion: The Ancient Roots of a Modern Concept. Style 2017, 51 (1), 34-51.

Anderson S. R. and E. L. Keenan. Deixis, in: T. Shopen (ed.), Language Typology and Syntactic Description.

Vol. 3: Grammatical Categories and the Lexicon. Cambridge, CUP, 1985, 259-308.

Bache C. English Tense and Aspect in Hallidays Systemic Functional Grammar: A Critical Appraisal and an Alternative. London, Equinox, 2008.

Bentein K. Aspectual Choice with Verba Dicendi in Herodotus' Histories. Emerita 2015, 83 (2), 221-245. Briscoe J. (ed. comm.). Titi Livi Ab Urbe Condita, Vol. 3: Libri XXI-XXV. Oxford, OUP, 2016.

Conway R. S. and S. K. Johnson (eds. comms.). Titi Livi Ab Urbe Condita, Vol. 4: Libri XXVI-XXX. Oxford, OUP, 1935.

Conway R. S. and C. F. Walters (eds. comms.). Titi Livi Ab Urbe Condita, Vol. 2: Libri VI-X. Oxford, OUP, 1919.

De Jong I. J. F. and R. Nunlist. From Bird's Eye View to Close-up: The Standpoint of the Narrator in the Homeric Epics, in: A. Bierl, A. Schmitt and A. Willi (eds.), Antike Literatur in neuer Deutung: Festschrift fur Joachim Latacz anlasslich seines 70. Geburtstages. Munchen -- Leipzig, Saur, 2004, 63-83.

Dorey T. A. (ed. comm.). Titi Livi Ab Urbe Condita Libri XXI-XXII. Leipzig, Teubner, 1971.

Fillmore C. J. Lectures on Deixis. Stanford, CA, CSLI Publications, 1997.

Fleischman S. Tense and Narrativity: From Medieval Performance to Modern Fiction. London, Routledge, 1990.

Foster B. O. (ed. comm.) Livy: In Fourteen Volumes. 1: Books I-II. Cambridge, Mass., HUP, 1919.

Foster B. O. (ed. comm.) Livy: In Fourteen Volumes. 3: Books V-VII. Cambridge, Mass., HUP, 1924.

Foster B. O. (ed. comm.). Livy: In Fourteen Volumes. 5: Books XXI-XXII. Cambridge, Mass., HUP, 1929. Gundel J. Information Structure and Referential Givenness/Newness: How Much Belongs in the Grammar?, in: S. Muller (ed.), Proceedings of the 10th International Conference on Head-Driven Phrase Structure Grammar, Michigan State University. Stanford, CA, CSLI Publications, 2003, 122-142.

Halliday M. A. K. and C. M. I. M. Matthiessen. Hallidays Introduction to Functional Grammar. Milton Park, Abingdon, Oxon, Routledge, ⁴2014.

Haverling G. V. M. Actionality, Tense, and Viewpoint, in: P. Baldi & P. Cuzzolin (eds.), New Perspectives on Historical Latin Syntax. Vol. 2: Constituent Syntax: Adverbial Phrases, Adverbs, Mood, Tense. Berlin -- New York, De Gruyter, 2010, 277-523.

Hope A. C. A. A Simplified Monte Carlo Significance Test Procedure. Journal of the Royal Statistical Society: Series B (Methodological) 1968, 30 (3), 582-98.

Kiss S. Tendances evolutives de la syntaxe verbale en latin tardif. Debrecen, Kossuth Lajos Tudomanyegye- tem, 1982.

McIntyre D. Point of View in Plays: A Cognitive Stylistic Approach to Viewpoint in Drama and Other Text- Types . Amsterdam -- Philadelphia, John Benjamins, 2006.

Mittwoch, A., R. Huddleston, and P. Collins. The Clause: Adjuncts, in: R. Huddleston and G. K. Pullum (eds.), The Cambridge Grammar of the English Language. Cambridge, CUP, 2002, 663-784.

Moore F. G. (ed. comm.). Livy: In Fourteen Volumes. 8: Books XXVIII -- XXX. Cambridge, Mass., HUP, 2002.

Ogilvie R. M. (ed. comm.). Titi Livi Ab Urbe Condita, Vol. 1: Libri I-V. Oxford, OUP, 1974.

Pinkster H. The Discourse Function of the Passive, in: A. M. Bolkestein, C. de Groot, and J. L. Mackenzie (eds.), Syntax and Pragmatics in Functional Grammar. Berlin -- New York, De Gruyter, 1985, 107-118.

Reichenbach H. Elements of Symbolic Logic. New York, Macmillan & Co, 1947.

Stockwell P. Cognitive Poetics: An Introduction. London, Routledge, 2002.

Thompson G. Introducing Functional Grammar. London, Routledge, ³20 14.

Tucker G. H. The Lexicogrammar of Adjectives: A Systemic Functional Approach to Lexis. London, Cassell, 1998.

Walsh P. G. (ed. comm.) Titi Livi Ab Urbe Condita Libri XXVIII-XXX. Leipzig, Teubner, 1986.

Размещено на Allbest.ru