From the article of F. Marx (‘Die Beziehungen des Altlateins zum Spätlatein’, Neue Jahrbücher 23/24 [1909], 434–48) onwards, ‘submerged Latin’ has been adopted as a general label in order to explain phenomena in early Latin which are absent in Classical texts, but prefigure a late Latin or Romance pattern. According to the standard view, submerged continuity originated in Latin's informal registers: sub-standard usages went underground during the Classical period (being prescriptively excluded from the literary language), but continued in the spoken language and resurfaced in the late period, when grammatical constraints were relaxed. However, as A. writes in Social Variation and the Latin Language (2013, p. 842), ‘scholars have sometimes blindly assigned phenomena to Vulgar Latin … for no better reason than that these phenomena influenced the Romance languages’. One can therefore wonder whether the ‘submerged Latin’ model is completely satisfying.
The book springs from a workshop, ‘Early Latin and Late Latin/Romance: Continuity and Innovation’, held at the University of Manchester on 12–13 May 2014, in order to test the model of ‘submerged Latin’ and more generally to identify patterns of continuity between early and late Latin. The sixteen contributions share a diachronic perspective in focusing mainly on structures which are attested both in pre-Classical usage and in the post-Classical era (and/or constitute the foundations of structures continued in Romance languages), so making a karstic development plausible.
The issues addressed concern vocabulary (G. Pezzini on comic words resurfacing in late Latin; J. Clackson's linguistic analysis of a series of Pompeiian graffiti) and morphology: on this subject – except T. Mari's chapter on the evolution of third person possessives – the main topic is the shift from synthetic to analytic forms, which is discussed by L. Danckaert, P. Burton and G. Haverling regarding verbal morphology, and by B. Bauer and R. Maltby regarding the comparative (and superlative) constructions.
But the field which requires thorough analysis is of course syntax. The contributions discuss the alternation between dative and ad + accusative (A. and W. de Melo); the impersonal temporal habet construction (S. Panayotakis); the use of the present indicative in deliberative questions (A. Chahoud). The development of infinitives complementing different types of verbs – inchoatives (G. Galdi), motion verbs (A. and V.) or causatives (V.) – is investigated thoroughly, while the last two chapters discuss dislocation: left-detached constructions (H. Halla-aho) and relative-correlative sentences (P. Probert and E. Dickey). Three pairs of contributions are partially overlapping in the choice of the topic, but different in methodology, approach and corpus selection: Danckaert and Burton, Bauer and Maltby, Halla-aho and Probert and Dickey.
This volume demonstrates that behind the apparent neatness of diachronic patterns lie issues of considerable complexity, raising several problems and further questions. Since the chapters adopt different approaches and different methodologies of data collection, a primary question concerns the selection of the corpus and the method of survey (e.g. quantitative vs qualitative). It is true, as A. states (p. 429), that ‘the more evidence the better if diachronic patterns of language change are to be identified’; however, where multifaceted phenomena such as linguistic variation are concerned, empirical data should be gathered and interpreted with great caution, since the effects of chronology cannot be clearly distinguished from those of co-varying factors such as genre or register, and the distribution of competing forms is not correlated with a single discriminating factor, but is rather the result of the complex interplay of a number of variables. Therefore, rough figures are mostly uninformative and require closer inspection. It is useful, for instance, to set a given usage against the background of other competing forms: this is the case in, among others, the contribution of A. and V., comparing data on infinitives of purpose with those on other purpose complements attested with motion verbs.
Contributions such as those of Pezzini and Danckaert make a case for the use of corpus data and statistical methods in studying linguistic phenomena, while Probert and Dickey reflect on the possibility of unconscious bias caused by corpus selection.
The labels traditionally used to identify varieties of Latin are a problematic matter. In his introductory chapter, V. speaks of ‘periodisation paradox’ in warning against over-reliance on chronological categories such as early, Classical or late Latin, and stresses the ambiguities of the label ‘vulgar Latin’, in which diachronic and sociolinguistic factors overlap. Indeed, in some contributions sociolinguistic variation, which represents a key factor, is taken less into account than diachronic development, confirming how difficult it is to assess valid sociolinguistic criteria in approaching written sources. A good sample of a corpus data analysis, which also pays attention to register and function, is given by Chahoud.
A further problem is posed by the relation between written sources and the living spoken language: as frequently stated in many chapters, the written language is also subject to fashions on a sub-literary level, or in genres having a habit of spoken language (such as comedy, speeches or, in the late period, homily), and therefore cannot offer direct evidence of patterns of spoken Latin. More subtly, Probert and Dickey suggest (p. 419) that the way Latin texts were transmitted might have been subject to biases in selection, so influencing our perception of linguistic phenomena.
In this respect, a more specific question arises, concerning the nature of Christian texts: do usages that entered Latin via Bible translations (this volume spots a number of instances) survive at a pure literary level (like in the case discussed by Galdi), or do they spread from there in the ordinary spoken language and influence its evolution (as A. and de Melo suggest happened with verbs of saying used with ad)? This is an important matter: striking cases of the influence of biblical usage on everyday Latin and on the Romance languages have been discussed already by E. Löfstedt in his classic work Late Latin (1959, pp. 81–7); but a particular biblical construction (‘translationese’) might have had no real life in the language, making the analysis of textual data difficult.
In conclusion, as far as ‘submerged Latin’ is concerned, the outcome is clear: essentially, every potential submerged feature taken into account does not stand up to scrutiny, and every contributor dealing with the question of whether a given phenomenon represents a true pattern of continuity between early and late Latin concludes – even if with different nuances and in a more or less explicit way – that it does not, or at least that it does so in a very partial manner.
‘Similarity is not continuity’, as A. and V. point out (p. 293): similarities between usages of the early period and of late Latin are often deceptive, since on one hand similar structures can arise for different reasons at different periods, and on the other hand, on closer inspection, can turn out not to be good matches syntactically and semantically, or can be due to learned borrowings or to independent recoinage.
Using a variety of methodological approaches and arguments, this book – thanks to the authority of its editors – calls into question a well-established model, thus reinvigorating a scholarly debate which will be of great interest in the future.
