Navigation – Plan du site
Perspectives diachroniques

Personal versus impersonal passive in Latin infinitival clauses: Some diachronic considerations

Passif personnel vs passif impersonnel dans les propositions infinitives latines : considérations diachroniques
Sean Gleason
p. 267-275

Résumés

Cet article traite les origines historiques de la concurrence, en latin classique, entre le NcI passif personnel et le passif impersonnel + AcI. Suite à l’observation faite par Maraldi (1983) que cette concurrence n’est pas attestée avant Cicéron au premier siècle avant notre ère, on soutient qu’une combinaison de facteurs linguistiques internes et externes se cache derrière son émergence dans la langue. Le facteur externe est l’existence d’une concurrence identique en grec ancien, tandis que le facteur interne est le développement du passif comme une catégorie indépendante en latin. L’interaction entre ces deux facteurs invite à envisager le passif impersonnel + AcI comme un hellénisme partiel.

Haut de page

Texte intégral

1. Introduction

1In the Accusativus cum Infinitivo (AcI) – illustrated in (1) – a matrix, or main, verb (e.g. dicit ‘he says’) introduces a complement clause containing an infinitive (venisse ‘to have come’) whose subject is marked in the accusative case (te ‘you’):

(1)

Dicit te venisse.
‘He says that you have come.’

  • 1 Pinkster, 1992, p. 163-4, defines impersonal passives as “subjectless passives”, and, pointing to t (...)

2When the matrix verb of (1) is passivized, two options are available: the personal passive, or Nominativus cum Infinitivo (NcI), seen in (2); or the impersonal passive,1 seen in (3):

(2)

Tu diceris venisse.
‘You are said to have come.’

(3)

Dicitur te venisse.
‘It is said that you have come.’

3In the personal passive (2) the subject of the infinitive now bears nominative case (tu ‘you’) and also serves as the subject of the matrix verb, as evidenced by the second-person agreement on diceris ‘you are said.’ Compare this with the impersonal passive (3), in which the subject of the infinitive retains its accusative case and does not also serve as the subject of the matrix verb dicitur ‘it is said’, which bears default third-person agreement.

  • 2 E.g. Hahn, 1950; Hettrich, 1992; Cuzzolin, 1994; Calboli, 2002.
  • 3 Cf. Hofmann-Szantyr, 1972, p. 815; Flobert, 1975, p. 466.
  • 4 See discussion in Calboli, 2005, p. 242f. Contra Ernout, 1908-9, p. 332; Woodcock, 1959, p. 22; a.o

4Diachronic studies of the AcI have not often been concerned with this alternation, instead largely focusing on the origins of the AcI itself and/or its relationship with quod-clauses in the development of Latin indirect speech.2 Yet it has not been completely ignored, particularly in Maraldi (1983, p. 174), where the observation is made that the use of the AcI with passive matrix verbs (the impersonal passive) is found only from Cicero in the first century BC onwards3 – it is only the NcI that is found in older Latin.4

5Now although this observation tells when the alternation between (2) and (3) came about, how and why it developed still remain to be answered. This paper serves as a starting point for doing just that by proposing that the change from the passive + NcI of older Latin to the passive + NcI/AcI alternation of Cicero onwards results from a combination of external and internal linguistic factors. The external factor is the influence of Ancient Greek, where the same alternation is found. The internal factors stem from the development of the Latin “passive” as an independent grammatical category, particularly with respect to its pragmatic function of either establishing or interrupting discourse continuity. This mix of internal and external factors points to viewing the change as a “partial Graecism,” or an internal development in Latin that is supported and/or triggered by a corresponding construction in Ancient Greek.

6To make this claim, section 2 provides a brief overview of the distribution of infinitival clauses in Early and Classical Latin, highlighting the spread of the AcI to passive matrix verbs. The Ancient Greek data is presented in section 3, while section 4 looks at the discourse differences between the two variants and how they fit into the historical development of the Latin passive. The conclusion follows, in which avenues for future research are considered.

2. Distribution of infinitival clauses in Early and Classical Latin

2. 1. Early Latin

  • 5 For periodization, see Cuzzolin and Haverling, 2009, p. 10.
  • 6 In these and the following examples, the AcI (set off with brackets) serves as the complement claus (...)

7In Early Latin (c. 3rd – early 1st centuries BC)5 both active (4) and deponent (5) matrix verbs take the AcI as a complement:6

(4)

Ubi dicimus [redisse te]. (Ter. Haut. 304)
‘When we said (lit. ‘say’) that you (had) returned.’

(5)

Id pollicetur [se daturum aurum mihi]. (Plaut. Bacch. 742)
‘He promises that he will give this gold to me.’

8In contrast, passive matrix verbs can only take the NcI in Early Latin, never the AcI (see Bennett, 1910, p. 388):

(6)

Is/*eum nunc dicitur ventur-us/*-um peregre. (Plaut. Truc. 84-5)
‘He is now said to be about to come from abroad.’

  • 7 Note that dicitur ‘he is said’ in (6) is an example of true third person agreement as it agrees wit (...)
  • 8 Note that this is based on preliminary corpus searches – more exhaustive corpus work is to come in (...)

9As noted above, the subject of the infinitive is also the subject of the matrix verb (is ‘he’ for dicitur ‘he is said’)7 and is only found attested in the nominative case – such subjects are never found in the accusative case, as denoted by the asterisks.8

2. 2. Classical Latin

10Just as in Early Latin, both active (10) and deponent (11) matrix verbs can take AcI complements:

(7)

Scio [me paene incredibilem rem polliceri]. (Caes. Civ. 3, 86, 2)
‘I know that I am promising an almost incredible thing.’

(8)

Non arbitror [te ita sentire]. (Cic. Fam. 10, 26, 2)
‘I do not think that you feel thus.’

11With respect to passive matrix verbs, they also follow their counterparts in Early Latin by taking the NcI:

(9)

Ad Themistoclem quidam doctus homo dicitur accessisse. (Cic. De orat. 2, 298)
‘A certain learned man is said to have approached Themistocles.’

12A marked change from Early Latin, however, is that in Classical Latin these passive verbs can also take the AcI as a complement:

(10)

Traditum est [Homerum caecum fuisse]. (Cic. Tusc. 5, 4)
‘It was related that Homer was blind.’

13It is this change that requires explanation.

3. External factor: the Greek connection

3. 1. The Ancient Greek data

14The same alternation exists in Ancient Greek between the impersonal and personal passives of indirect speech:

(11)

Σοὶ γὰρ δὴ λέγεται πάνυ γε τεθεραπεῦσθαι Ἀπόλλων, καί σε πάντα ἐκείνῳ πειθόμενον πράττειν. (Xenoph. Cyr. 7, 2, 15)
‘For Apollo is said to have been greatly served by you, and (it is said) that you do everything in obedience to him.’

15The first half of the conjunct σοὶ Απόλλων contains the NcI, where the matrix verb λέγεται ‘is said’ takes a clausal complement containing the infinitive τεθεραπεῦσθαι ‘to have been served’, and the nominative Απόλλων ‘Apollo’ is the subject of both of these verbs – this is identical to the Latin NcI.

16Contrast this with the second half of the conjunct σεπράττειν, which contains an impersonal passive in which the understood matrix verb λέγεται ‘it is said’ takes a clausal complement containing the infinitive πράττειν ‘to do’ whose subject is the accusative σε ‘you’. Note that σε ‘you’ is the subject of the infinitive only and not of the understood matrix verb, just as is in the relevant Latin construction.

3. 2. The Latin impersonal passive: a Graecism?

  • 9 Cf. Poccetti et al., 1999, p. 112, and the discussion and references in Calboli 2009, p. 65-77.

17The data in (11) points to Ancient Greek as the source of the Latin impersonal passive + AcI construction. This would treat the impersonal construction, along with it resultant alternation with the personal passive NcI, as a Graecism, or a borrowing/imitation of a Greek construction in Latin. Yet before settling on this as the sole explanation for the situation in Latin, it must be noted that teasing out the full extent of Graecisms in Latin syntax is often complicated by the possibility that a given construction can be as much an independent or parallel development in Latin as it can be the result of Greek influence.9 This possibility must give pause to labelling a given Latin construction as a Graecism, as failure to take it into consideration may erroneously exaggerate the influence of Greek upon the language (e.g. Brenous, 1895).

18However, in seeking to avoid exaggerating the presence of Graecisms in Latin syntax it is equally important to avoid diminishing them. For this reason it is worth considering the notion of a “partial Graecism,” first introduced in Löfstedt (1933, p. 416), and elaborated in Coleman (1975, p. 103-4), which Calboli (2009, p. 69), succinctly defines as “a natural development within Latin that is reinforced by a similar construction in Greek.” This is echoed in Rósen (1999, p. 26), where she claims that, in general, “Graecism in syntax, while mostly beginning as literary imitatio, could trigger systemic changes and produce new phenomena, provided they met with language features existing before the ‘argentea’.” This notion sees interplay between Greek syntactic influence and Latin’s own internal linguistic developments, making the relationship between them more than just a zero-sum game between Greek borrowings and independent Latin innovations. Identifying these partial Graecisms is thus a way of acknowledging the influence Greek had on Latin syntax without having to exaggerate or ignore it.

19This also suggests that determining the role of Ancient Greek in the emergence of Latin syntactic phenomena must first consider what internal diachronic changes, if any, might have existed in the latter to facilitate borrowing from the former. That is, identifying the influence of Greek on Latin syntax is not as simple as saying whether or not a given construction comes from Greek, but must also take into account whether some relationship exists between that construction and the broader diachronic developments of Latin itself. The next section of the paper deals with just that.

4. Internal factors

4. 1. Development of the Latin “passive”

  • 10 See Fortson, 2004, p. 82; Clackson, 2007, p. 142; Bauer, 2000, p. 238; a.o.

20The internal factors involved in the emergence of the Latin impersonal construction center on the development of the passive into a full and independent grammatical category within the language. The original voice distinction in Proto-Indo-European (PIE) seems to have been between active and medio-passive (MP)10 – this is supported by some of the earliest attested Indo-European languages, particularly Hittite and Ancient Greek. While independent active and middle forms can be reconstructed for PIE, the evidence is less strong for a specialized passive –instead it more robustly points to passive uses of the middle form, hence the label medio-passive (cf. Kulikov and Lavidas, 2013, p. 102-3 and the citations contained therein). The development of a fully independent passive grammatical category likely occurred later within the daughter languages, they themselves having originally inherited the PIE active/MP distinction.

21This development of an inherited MP voice into an independent passive category is true of Latin, a fact noted from Allen and Greenough (1903, p. 73) to Clackson and Horrocks (2011, p. 25). Rosén (1999, p. 124-37) claims they originated in sentences like (12), where the agent is not in focus:

(12)

Peccatum est a me. (Ter. Haut. 158)
‘The mistake was made by me.’

  • 11 Original fast unverändert.
  • 12 Note that this amounts to claiming that a passive in writers like Plautus or Terence is not a passi (...)

22Chronologically the development must have largely occurred early in the history of Latin, as Calboli (1990, p. 110) sees the passive voice as a fully established semantic category that remains “nearly unchanged” from Archaic to Classical Latin.11 However, as demonstrated by Calboli’s qualification and by the discussion in Rosén (1999, p. 124-37), the passive was to undergo several more developments from the earlier to later periods of the language before a full active-passive voice distinction would exist in Latin. It would not be until Classical Latin that the passive had acquired all of the morpho-syntactic, semantic, and pragmatic properties usually associated with it.12

4. 2. Personal/impersonal distinction

  • 13 Cf. Pinkster, 1990, p. 131; Kühner and Stegmann, 1912-14, p. 707-9.

23This last point is of most relevance for the passive + AcI construction, but, before elaborating further, it is necessary to discuss an important pragmatic difference between it and the NcI. Bolkestein (1983, p. 121-2) notes that in the personal passive NcI some element of the infinitival clause, usually the subject, is in focus:13

(13)

et multis ante saeculis Lycurgum, cuius temporibus Homerus etiam fuisse ante hanc urbem conditam traditur. (Cic. Tusc. 5, 7)
‘… and Lycurgus many ages before, in whose time before the founding of the city
Homer is also held to have lived.’

  • 14 This contrast is further expressed through the presence of the focusing particle etiam ‘also’.

24In (13) the focus is on the nominative-marked Homerus ‘Homer’, who is contrasted as subject with the previously mentioned Lycurgus.14 Compare this with the passive + AcI, or impersonal passive, in which the statement expressed by the infinitival clause is related without focusing on the individual element of its subject (cf. Bolkestein, 1983, p. 121-2):

(14)

Eorum una, pars, quam Gallos obtinere dictum est, initium capit a flumine Rhodano […] vergit ad septentriones. (Caes. Gal. 1, 1, 5)
One part of these, which it was said that the Gauls occupy, takes its beginning at the River Rhone and stretches towards the north.’

25In (14) the subject of the infinitival clause is the accusative Gallos ‘the Gauls’, but this subject is not in focus. What is being talked about in the larger discourse is one of the three parts into which Gaul is divided (eorum una, pars ‘one of these, a part’), and the use of the AcI in quam Gallos obtinere dictum est ‘which it is said that the Gauls occupy’ allows the author to provide further information with a different subject (i.e. Gallos) without shifting the focus away from eorum una. In other words, by marking Gallos in the accusative it lets the reader know that although this is a different subject, Caesar is not shifting focus away from the earlier mentioned subject eorum una, pars.

4. 3. Creating the distinction

26The pragmatic effect of the accusative Gallos ‘the Gauls’ in (14) is thus very different from the nominative Homerus ‘Homer’ in (13). The use of the nominative subject requires that this subject be put into focus, usually in contrast to a previous, different subject; the use of the accusative subject allows for the its mention without having to take the focus off of a previous, different subject. This is in line with what Pinkster (1985, p. 114-15) notes as one of the discourse functions of the passive: it can serve as either a continuity or discontinuity device with respect to the subject of the sentence. That is, one of the functions of the passive is that it can be used to continue speaking about the same subject in discourse or break continuity by bringing a new subject into focus.

27It is the very expansion of AcI complements to passive matrix verbs that creates the possibility for the contrast between (13) and (14). This extension allows for an option in passive constructions between putting the infinitival subject in focus (the NcI) or not (AcI) – i.e. there is an increase in discourse strategies available to Latin following the development of the impersonal construction. Put another way, this change marks an increase in the pragmatic properties of the passive from Early to Classical Latin, which fits with the fact discussed in section 4.1 that it is only in the shift between these two diachronic periods that the Latin passive comes to fully acquire all of the morpho-syntactic, semantic, and pragmatic properties usually associated with it. The emergence of the passive + AcI should thus be understood as one of the developments that results in the passive becoming a full and independent grammatical category in Latin.

28By extending the AcI complement to passive matrix verbs and having this exist alongside the NcI, Latin expands its use of the passive so as to be able to move between putting in focus a constituent of the infinitival clause, the infinitival clause itself, or something else entirely. In so doing, the ability to maintain continuity or introduce discontinuity with respect to something like the subject of discourse becomes an important pragmatic property of the Latin passive. In developing the two variants the Latin passive comes to more fully realize its pragmatic potential, and it is this increase in pragmatic properties, understood in light of the overall development of the Latin passive, that is at the center of the internal factors involved in the emergence of the impersonal construction.

5. Conclusion

29This paper began with the established observation that the Classical Latin alternation between the personal passive NcI and the impersonal passive with the AcI is first found in the writings of Cicero in the first century BC – prior to this, Latin passive matrix verbs only took the NcI as a complement. The emergence of the passive + AcI construction was attributed to a combination of external and internal linguistic factors, the external being the existence of an identical construction in Ancient Greek – this points to a syntactic borrowing into Latin. Yet given the complications involved in teasing out the exact nature of Greek influence on Latin syntax, this was not taken to be the sole explanation. The existence of “partial Graecisms” in particular raised the necessity for determining what, if any, internal factors might also be at work.

30The search for internal factors led to an examination of the development of a fully independent passive in Latin. Like other IE languages, Latin inherited an active/MP distinction from PIE and only over time developed an independent passive voice in opposition to the active. This development largely occurred early in the history of the language, but the final steps took place in the shift from Early to Classical Latin, during which the passive fully acquired the grammatical properties usually associated with it. One such pragmatic property was the passive’s functioning as a device for maintaining or disrupting discourse continuity, particularly with respect to subjects. The introduction of the passive + AcI as a variant alongside the NcI introduced the possibility of having something other than an individual constituent of the infinitival clause in focus, such as the infinitival clause itself or something else entirely, and in doing so increased the strategies for maintaining or disrupting continuity. The emergence of the impersonal construction was therefore understood as part of the expansion of an important pragmatic property within the development of the Latin passive from Early to Classical Latin.

  • 15 A reviewer asks whether it may also find support in constructions involving verbs with illocutionar (...)

31The combination of this internal development of the passive with the external influence of Greek points to the passive + AcI construction as a partial Graecism, but identifying it as such is one thing, determining causality is another. Following the definition of partial Graecism given above, the internal development of the Latin passive would have been supported by an identical construction in Greek. The extension of the AcI complement to passive matrix verbs would have initially been part of the internal development of the Latin passive, and its taking root as a regular construction in the language would have been reinforced by the existence of its Greek counterpart.15

32But it is also possible that the Greek construction is the trigger for the Latin development – i.e. the borrowing of the Greek construction spurs the ongoing development of the Latin passive to include the ability to take AcI complements and expand its pragmatic properties. Another distinct possibility is that the internal developments in Latin support the borrowing from Greek in that they provide a way for the borrowing to be integrated into the language as a regular construction rather than becoming a calque. Teasing out the exact interplay between the external and internal factors behind the change will have to remain the focus of subsequent work on this topic.

33It is necessary to end by emphasizing that this paper represents the beginnings of an investigation into the origins of the co-existence of the personal/impersonal variants in Latin infinitival clauses. The conclusions presented here are therefore preliminary pending further research. Of primary importance is more corpus work, beginning with an exhaustive corpus search of Early Latin. Particular issues to be addressed include what difference, if any, exists between Plautus and Terence, especially in light of the latter’s stronger dependence on Greek sources. Additional corpus work in Classical Latin will take a closer look at the contexts in which the impersonal passive construction appears, focusing on whether different genres of text affect the frequency of its use. This future work stands to both refine the claims presented here as well as add to our continuing understanding of the diachronic development of Latin syntax.

Haut de page

Bibliographie

Allen, J. H. and Greenough, J. B., 1903, New Latin Grammar, Boston.

Bauer, B., 2000, Archaic Syntax in Indo-European, Berlin.

Bennett, C., 1910, Syntax of Early Latin, vol. 1: The Verb, Boston.

Bolkestein, A. M., 1983, The role of discourse in syntax: Evidence from the Latin Nominativus cum infinitivo, in K. Ehlich and H. van Riemsdijk (eds.), Connectedness in Sentence, Discourse, and Text, Tilburg, p. 111-140.

Brenous, J., 1895, Étude sur les hellénismes dans la syntaxe latine, Paris.

Calboli, G., 1990, Vulgärlatein und Griechisch in der Zeit Trajans, in G. Calboli (ed.), Latin vulagire – Latin tardif II, Tübingen, p. 23-44.

Calboli, G., 2002, Moods and tenses in declarative clauses: At the origin of the AcI construction, in G. Calboli (ed.), Papers on Grammar 8, Rome, p. 69-89.

Calboli, G., 2005, Accusative as default case in Latin, IF 110, p. 235-266.

Calboli, G., 2009, Latin syntax and Greek, in Ph. Baldi and P. Cuzzolin (eds.), New Perspectives on Latin Historical Syntax, vol. 1: The Syntax of the Sentence, Berlin, p. 65-193.

Clackson, J. and Horrocks, G., 2011, The Blackwell History of the Latin Language, Malden, MA.

Clackson, J., 2007, Indo-European Linguistics: An Introduction, Cambridge.

Coleman, R., 1975, Greek influence on Latin syntax, TPhS 74 (1), p. 101-56.

Cuzzolin, P. and Haverling, G., 2009, Syntax, sociolinguistics, and literary genres, in Ph. Baldi and P. Cuzzolin (eds.), New Perspectives on Latin Historical Syntax, vol. 1: The Syntax of the Sentence, Berlin, p. 19-64.

Cuzzolin, P., 1994, Sull’origine della costruzione ‘dicere quod’: Aspetti sintattici e semanticii, Florence.

Ernout, A. and Thomas, F., 1953, Syntaxe latine, Paris (2nd edition).

Ernout, A., 1908-9, Recherches sur l’emploi du passif latin à l’époque républicaine, Mémoires de la Société de Linguistique de Paris 15, p. 273-333.

Flobert, P., 1975, Les verbes déponents latins : des origines à Charlemagne, Paris.

Fortson, B., 2004, Indo-European Language and Culture: An Introduction, Malden, MA.

Gleason, S., in preparation, The Interaction of Case and Voice: Insights from the Latin Accusativus cum Infinitivo, PhD thesis, Yale University.

Hahn, E. A., 1950, Genesis of the infinitive with subject-accusative, TAPhA 81, p. 117-129.

Hartman, J., 2012, Varieties of Clausal Complementation, PhD thesis, Massachusetts Institute of Technology.

Hettrich, H., 1992, Die Entstehung des lateinischen und griechischen AcI, in R. Beekes, A. Lubotsky, and J. Witenberg (eds.), Rekonstruktion und relative Chronologie, Innsbruck, p. 221-234.

Hofmann, J. B. and Szantyr, A., 1972, Lateinische Syntax und Stilistik, Munich.

Kühner, R. and Stegmann, C., 1912-1914, Ausführliche Grammatik der Lateinischen Sprache, Hannover.

Kulikov, L. and Lavidas, N., 2013, Reconstructing passive and voice in Proto-Indo-European, Journal of Historical Linguistics 3 (1), p. 98-121.

Löfstedt, E., 1933, Syntactica: Studien und Beiträge zur historischen Syntax des Lateins, vol. 2, Lund.

Maraldi, M., 1983, New approaches to accusative subjects: Case theory vs. raising, in H. Pinkster (ed.), Latin Linguistics and Linguistic Theory, Amsterdam, p. 167-176.

Pinkster, H., 1985, The discourse function of the passive, in A. M. Bolkestein, C. de Groot, and J. L. Mackenzie (eds.), Syntax and Pragmatics in Functional Grammar, Dordrecht, p. 107-118.

Pinkster, H., 1990, Latin Syntax and Semantics, London.

Pinkster, H., 1992, The Latin impersonal passive, Mnemosyne 45, p. 159-177.

Poccetti, P., Poli, D., and Santini, C., 1999, Identità e identificazione del latino, in P. Poccetti, D. Poli, and C. Santini (eds.), Una storia della lingua latina, Rome, p. 9-171.

Rosén, H., 1999, Latine loqui: Trends and Directions in the Crytallization of Classical Latin, Munich.

Woodcock, E. C., 1959, A New Latin Syntax, Cambridge, MA.

Haut de page

Notes

1 Pinkster, 1992, p. 163-4, defines impersonal passives as “subjectless passives”, and, pointing to the traditional analysis of the AcI clause itself serving as the subject of dicitur in (3), thus does not consider this as properly impersonal. However, this clause lacks several of the formal properties shared cross-linguistically by clauses that serve as subjects of a sentence, such as the licensing of emphatic reflexives (cf. Hartman, 2012, p. 44-55), calling its subject-status into question. In light of this, I follow Hofmann and Szantyr, 1972, p. 358-9 in referring to (3) as an impersonal passive.

2 E.g. Hahn, 1950; Hettrich, 1992; Cuzzolin, 1994; Calboli, 2002.

3 Cf. Hofmann-Szantyr, 1972, p. 815; Flobert, 1975, p. 466.

4 See discussion in Calboli, 2005, p. 242f. Contra Ernout, 1908-9, p. 332; Woodcock, 1959, p. 22; a.o.

5 For periodization, see Cuzzolin and Haverling, 2009, p. 10.

6 In these and the following examples, the AcI (set off with brackets) serves as the complement clause to the respective active matrix verb (highlighted in bold).

7 Note that dicitur ‘he is said’ in (6) is an example of true third person agreement as it agrees with the third person subject is ‘he’. This is in contrast with the dicitur ‘it is said’ of (3), where there is no third person subject with which to agree, meaning its third person agreement marking must be taken to be default agreement.

8 Note that this is based on preliminary corpus searches – more exhaustive corpus work is to come in future work (Gleason, in preparation).

9 Cf. Poccetti et al., 1999, p. 112, and the discussion and references in Calboli 2009, p. 65-77.

10 See Fortson, 2004, p. 82; Clackson, 2007, p. 142; Bauer, 2000, p. 238; a.o.

11 Original fast unverändert.

12 Note that this amounts to claiming that a passive in writers like Plautus or Terence is not a passive in exactly the same way it is for someone like Cicero. This sentiment is also found in Ernout and Thomas, 1953, p. 201-8 (thanks to Dominque Longrée for bringing this to my attention).

13 Cf. Pinkster, 1990, p. 131; Kühner and Stegmann, 1912-14, p. 707-9.

14 This contrast is further expressed through the presence of the focusing particle etiam ‘also’.

15 A reviewer asks whether it may also find support in constructions involving verbs with illocutionary force, such as tu iuberis venire ‘you are ordered to come’. The issue is if there exists an impersonal variant, such as iubetur te venire ‘it is ordered that you come’ in the Early period that might serve as an analogical source for the dicitur te venire construction. Based on preliminary corpus searches, I cannot find an instance of iubetur te venire attested, which would suggest it is not a source for the impersonal dicitur construction. If there is anything in Latin that might serve as an analogical source/influence, it might be the formally identical deponent verbs that take AcI complements throughout the history of Latin, but the details of a claim like this will have to be addressed in future work.

Haut de page

Pour citer cet article

Référence papier

Sean Gleason, « Personal versus impersonal passive in Latin infinitival clauses: Some diachronic considerations », Pallas, 102 | 2016, 267-275.

Référence électronique

Sean Gleason, « Personal versus impersonal passive in Latin infinitival clauses: Some diachronic considerations », Pallas [En ligne], 102 | 2016, mis en ligne le 20 décembre 2016, consulté le 17 décembre 2017. URL : http://journals.openedition.org/pallas/3751 ; DOI : 10.4000/pallas.3751

Haut de page

Auteur

Sean Gleason

PhD student, Yale University
sean.gleason@yale.edu

Haut de page

Droits d’auteur

Licence Creative Commons
Pallas – Revue d'études antiques est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo Presses universitaires du Midi
  • OpenEdition Journals