Home Romanization and Latinization of the Roman Empire in the light of data in the Computerized Historical Linguistic Database of Latin Inscriptions of the Imperial Age
Article Open Access

Romanization and Latinization of the Roman Empire in the light of data in the Computerized Historical Linguistic Database of Latin Inscriptions of the Imperial Age

  • Béla Adamik EMAIL logo
Published/Copyright: June 7, 2021
Download Article (PDF)
Become an author with De Gruyter Brill
Author Information
Journal of Latin Linguistics
From the journal Journal of Latin Linguistics Volume 20 Issue 1

Abstract

The present study demonstrates that the process of linguistic Romanization, i.e. Latinization of the Roman Empire, is traceable by the data of the Computerized Historical Linguistic Database of Latin Inscriptions of the Imperial Age (LLDB). A multi-level analysis of linguistic and non-linguistic data in the LLDB has shown that Latinization, i.e. the spread of spoken or vulgar Latin, became more and more intensive over time in all concerned provinces (i.e. Lusitania, Gallia Narbonensis, Venetia et Histria, Dalmatia, Moesia, Pannonia, and Britannia), although to a varying degree in each. What is more, in many aspects of the investigation, it was possible to find differences between the selected provinces of the Roman Empire corresponding mostly to the future Romance (both negative and positive) outcomes of the respective areas. All in all, the analysis of data of the LLDB database can contribute to solving the complex problem of Latinization, and is a lot more appropriate for this purpose than a simple comparative analysis of epigraphic corpora of the selected provinces.

Keywords: database; inscriptions; Latinization; Romanization; vulgar Latin

1 Introduction

For contributing to the discussion on the challenging and still somehow controversial problem of the linguistic Romanization, i.e. Latinization[1] of the Roman Empire, I am going to involve the data of the Computerized Historical Linguistic Database of Latin Inscriptions of the Imperial Age (“the Database” or “LLDB”[2] henceforth). The aim of this project is to develop and digitally publish (at http://lldb.elte.hu/) a comprehensive, computerized historical linguistic database that contains and manages the Vulgar Latin material of the Latin inscriptions found in the different territories of the Roman Empire (Illyricum, Gallia, Britannia, Germania, Hispania, Italia, Africa, Roma and the eastern provinces). This will allow for a more thorough study of the diatopic changes and variation of the Latin language of the Imperial Age in a wider sense, and for a multilayer visualization of the structures discovered concerning linguistic geography.[3]

In this paper, nevertheless, I will not be dealing with actual dialectological problems – however vital and interesting they are – but with more general problems related to the linguistic Romanization and Latinization of the Roman Empire. By that I mean the spread of spoken or vulgar Latin (preceded and triggered by mostly spontaneous language shift[4] through the agency of the vernacular populations) both territorially and chronologically.[5] These aspects will be addressed in light of data in the Database. The presentation of the problem and the research questions of my paper are as follows: Is the process of Latinization of the Empire traceable by the data of such a linguistic database? If yes, can any differences be evidenced between the provinces of the Roman Empire? If there are differences to be revealed between the provinces, is there any correlation between these differences and the future Romance continuation or discontinuation in the respective provinces, that is, whether Romance languages developed there or not? Can such a linguistic database contribute to this complex problem?

2 Methodology

To answer these difficult questions we shall use the Database, which has some unique features. First of all, its entries consist of deviations of all sorts from “Classical Latin” or what is called the “classical norm,”[6] irrespective of whether they are due to linguistic or extra-linguistic (i.e. graphic or technical, etc.) factors. LLDB data collectors record not only data of linguistic but also of technical origin; moreover, we also differentiate between actual linguistic data and purely orthographic phenomena. We have accumulated a huge number of data forms in the Database (more than 100,000),[7] which means we can work with exact, high figures and accurate rates for every relevant aspect related to the linguistic Romanization and Latinization of the Roman Empire. This means we can analyze the spread of spoken or vulgar Latin – as opposed to previous work on the subject, which instead described the general conditions and course of this process, without a detailed analysis of the specific Latin linguistic data.[8]

As for the territorial delimitations of my current analysis, the following restrictions must be taken into consideration. First, although the Database contains data from the entire territory of the Empire, now only the western or the so-called Latin part of the Empire is considered, while the eastern or so-called Greek part is excluded.[9] The reason for this is that the Greek part had quite different sociolinguistic peculiarities that were unfavorable for a linguistic Romanization, i.e. Latinization.[10] Secondly, from the vast territory of the western or Latin part of the empire, I selected only a few provinces, which are all well processed in the database project, that is, their epigraphic material is almost fully entered in the Database, so we have a considerable amount of data available from there. Regarding the Romance continuation or discontinuation, for areas characterized by a persistent Romance continuation I selected (i) Lusitania, (ii) Gallia Narbonensis and (iii) Venetia et Histria; then, for areas characterized by an extinct Romance continuation I selected (iv) Dalmatia;[11] and, finally, for areas of no romance continuation, I selected (v) Moesia, (vi) Pannonia, and (vii) Britannia, where no Romance language evolved due to the known historical and sociolinguistic circumstances.[12] In order to see the changes over time, I divided the relevant material into two periods: an early one from the 1st through the 3rd century, and a later one from the 4th through the 7th century (see Figures 1a and 1b).[13]

Figure 1a: Charts 1a–4b.
Figure 1a:

Charts 1a–4b.

Figure 1b: Charts 5a–8b.
Figure 1b:

Charts 5a–8b.

For the next step, we divided the data of both periods of each province into the following four categories: Phonologica and Morphosyntactica, representing the “linguistic” data (abbreviated as L in the chart titles of Figures 1a and 1b), and Orthographica and Errores non grammatici, representing the “non-linguistic” data (abbreviated as NL in the chart titles of Figures 1a and 1b). For each category, two (or in one case, three) illustrative examples are given directly below.

3 Analysis of linguistic and non-linguistic data

The linguistic data are illustrated by the following examples:[14]

  1. Phonologica:

    1. LLDB-7188: (voc.)-v-(voc.) > B, MILIT|ABIT = militavit, RIU 2, 559, 1–2 = CIL 3, 11026, 1–2, Pannonia superior, Brigetio, 301–350 AD.

    2. LLDB-11849: é: > I, POSVIRVNT = posuerunt, InscrAqu 3, 2,910, 6 = CIL 5, 1,680, 6, Venetia et Histria / Regio X, Aquileia, 365 AD.

  2. Morphosyntactica:

    1. LLDB-25589: acc. pro abl., CVM FILIOS = cum filiis, ILGN 102, 3 = CIL 12, 5,419, 3, Gallia Narbonensis, ?, 313–700 AD.

    2. LLDB-17177: commut. in form. perfecti activi verbi ponendi, POSIT = posuit, Conrad 268, 7 = AE 1972, 523, 7, Moesia inferior, Tropaeum Traiani, 151–200 AD.

The non-linguistic data are illustrated by the following examples:

  1. Orthographica: [15]

    1. LLDB-30272: H > ø, IC = hic, CIIC 1, 327, 1 = IBC 34, 1, Britannia, by Gobannium, 501–600 AD.

    2. LLDB-30361: x > SX / CS / XS / XSS / XX , VICSIT = vixit, CIL 3, 2,151, 6, Dalmatia, Salona, 151–230 AD.

    3. LLDB-60209: ae > E, SVE| = suae, AE 2013, 795, 7, Lusitania, by Norba, 190–310 AD.

  2. Errores non grammatici:

    1. LLDB-20923: litterae perperam incisae, EILIAE = filiae, InscrIt 10, 1, 559, 1, Venetia et Histria / Regio X, Pola , 313–600 AD.

    2. LLDB-31256: syllaba perperam adiuncta (dittographia), AGVSTA|STAS = Augustas, Badajoz 41, 9–10 = HEp 7, 154, 9–10, Lusitania, by Emerita Augusta, AD 574 AD.

In Figures 1a and 1b, I present data from each selected province in this fourfold distribution. In the first column you can see data from the early period (i.e. from the 1st through the 3rd century), and in the second column you can see data from the later period (i.e. from the 4th through the 7th century). In the heading of each chart of Figures 1a and 1b, after the abbreviated name of the Database (“LLDB”), you can see the name of the province, the time span of the charted data (e.g. Chart 1a in Figure 1a has “Lusitania c. 1–3 AD”), and between brackets the total number of data forms charted (ND, e.g. 1,674), the number of inscriptions with misspellings (NI, e.g. 990) since one inscription can often yield multiple data forms, and, accordingly, the average number of misspellings or deviations per inscription which we are going to refer to as Data density (ND/NI = 1.7). These figures are followed by the rate of linguistic data (L, e.g. 66 %) and the rate of non-linguistic data (NL 34 %). In the last charts of Figure 1b, 8a–8b, you can see displayed these phenomena for all provinces considered, with the average numerical values to be taken as a point of reference. This way the charts in Figures 1a and 1b, all generated by the charting module of the LLDB Database, represent the basic data sets for our interpretation to be explained right now.[16]

In Figures 2a–2f you can see charts interpreting and summarizing the data displayed in Figures 1a and 1b. They will help us compare and rank the selected provinces with respect to the chosen categories, discover the tendencies of changes between the early and the later period, and draw our conclusions with regard to Latinization, so we can see if there is any correlation between the data and the fate of the local Romance languages.

3.1 Analysis according to the number of inscriptions with misspellings

Let us start with Charts 1a–1c of Figure 2a, which display the number of registered inscriptions with misspellings, that is, containing any kind of deviations.

Figure 2a: Charts 1a–1c.
Figure 2a:

Charts 1a–1c.

From the distribution in the early period, displayed in Figure 2a, Chart 1a, we cannot infer the future Romance developments at all, since e.g. Pannonia with no Romance continuation takes the third place in the leading trio together with Gallia Narbonensis with Romance continuation and with Dalmatia with an extinct Romance continuation, while Lusitania and Venetia et Histria, both with a Romance continuation, are positioned in the group lagging behind, together with Moesia and Britannia, both with no Romance continuation. As opposed to the early period, the distribution in the later period, displayed in Figure 2a, Chart 1b, projects the future Romance developments in an obviously much more faithful way, with a clear split into two prominent groups, one with quite a huge number of registered inscriptions (ca. 300–500) including Gallia Narbonensis, Venetia et Histria, Dalmatia, and Lusitania, all of them with a Romance continuation; and one underdeveloped group with a low number of registered inscriptions (ca. 50–100) including Britannia, Moesia, and Pannonia, all with no Romance continuation. Only the position of Dalmatia does not show any correlation with the later fate of the local Romance language: with an extinct Romance continuation, we expected it somewhere halfway between the two groups with a number of ca. 200 entered inscriptions. Figure 2a, Chart 1c, displaying the difference in the number of inscriptions with misspellings between the early and the later period, does not help much regarding this issue, since even though Dalmatia holds the second worst position (with a decrease of 1,344 inscriptions) preceding Pannonia in the worst position (with a decrease of 1,361 inscriptions), Gallia Narbonensis got in a similar – that is, the third worst – position (with a decrease of 1,069 inscriptions), while Britannia (with a decrease of 522 inscriptions) got in the leading trio together with Lusitania and Venetia et Histria (with a decrease of 665 and 374 inscriptions, respectively).

3.2 Analysis according to the number of data forms

The data are better understood if we regard the distribution of the number of data forms instead of the number of inscriptions with misspellings in the selected provinces, as displayed in Figure 2b, Charts 2a–2c.

Figure 2b: Charts 2a–2c.
Figure 2b:

Charts 2a–2c.

While in the early period (cf. Figure 2b, Chart 2a), unsurprisingly, there is no correlation to be detected between the figures for each province and any future Romance development, in the later period a clear correlation is visible (see Figure 2b, Chart 2b). In the underdeveloped group we see Moesia, Britannia, and Pannonia, all with no Romance continuation (with figures between 100 and 300), and in the well-developed group we see Venetia et Histria, Lusitania, and Gallia Narbonensis, all with a persistent Romance continuation (with figures between 1,400 and 1,800), while Dalmatia with an extinct Romance continuation takes a sort of middle position, a bit lagging behind the developed group (with 1,112 data forms). In this case, Figure 2b, Chart 2c also reflects the future Romance developments in the difference of number of data forms between the early and the later period: the three provinces with a persistent Romance continuation are in the first three positions (Gallia Narbonensis with a decrease of only 549 items, Lusitania with a decrease of 75 items, and Venetia et Histria with an increase of 142 items). The two provinces with no Romance continuation are in the last and third from last positions (Pannonia with a decrease of 2,759 items and Moesia with a decrease of 1,686 items). At the same time, quite “disturbingly,” Dalmatia has a position in the middle of the latter group, with a huge decrease in data forms (2,084 items), while, by contrast (with a decrease of 707 items), Britannia gets very close to the developed group. However, Britannia’s distinguished position regarding both the decrease in the number of inscriptions with misspellings (Figure 2a, Chart 1c) and in the number of data forms (Figure 2b, Chart 2c) is rather illusory and misleading, as it can instead be explained by a superficial Latinization of the province in the early Imperial Age, resulting in low figures (actually the lowest figures in comparison to Charts 1a and 1b) both in the number of inscriptions with misspellings and of data forms.[17] In this case, the distributional patterns displayed in Figure 2b, Chart 2b match the distribution in Romance the best.

3.3 Analysis according to data density

Now we turn to Figure 2c, Charts 3a–3c, which display data density in the two periods and the changes in data density between the early and the later period.

Figure 2c: Charts 3a–3c.
Figure 2c:

Charts 3a–3c.

As I have mentioned earlier, data density here refers to the average number of misspellings or deviations per inscription. For example, in early Dalmatia we have 3,196 data forms (ND, cf. Chart 4 of Figure 1a and Chart 2a in Figure 2b), recorded from 1,717 inscriptions (NI, cf. Chart 4 in Figure 1a and Chart 1a in Figure 2a). If we divide the 3,196 data forms by the 1,717 inscriptions, we get a data density of 1.9 (rounded off to the first decimal place, see Chart 3a in Figure 2c). While in the early period (cf. Chart 3a in Figure 2c) again there is no correlation to be detected between the data of the provinces and the future Romance developments, in the later period a partial correlation is to be noticed (cf. Chart 3b in Figure 2c), even if not so clear as in Chart 2b (Figure 2b). For better comparison, in Charts 3a–3c (Figure 2c) we introduced an imperial average figure for “all provinces” (in Chart 3b in Figure 2c shown as 32,852/11,358 = 2.9, cf. Chart 8b in Figure 1b). If we measure the provincial data density figures against this imperial average figure (2.9), in the later period we can see that Britannia is set clearly below it (with 2.2 by 0.7), and Moesia clearly above it (with 3.5 by 0.6), while Dalmatia, Venetia et Histria, and Pannonia all show roughly the imperial average value (with 3, with a deviation of 0.1 %). This way only Britannia (2.2), Gallia Narbonensis (3.9), and Lusitania (4.9) seem to be in their appropriate places on the scale of data density in the later period. However, if we move on to the changes in data density between the early and the later period (Chart 3c in Figure 2c), we notice two significant phenomena. First, the data density increased both on the imperial average (1.1) and in all provinces concerned. If we consider the pattern of growth rates, we see a clear correlation between this pattern and the future Romance linguistic geographical distribution. Britannia with no Romance continuation has the lowest growth rate (0.7), then Pannonia with no Romance continuation has the second lowest growth rate (0.9), and Dalmatia with an extinct Romance continuation has the third lowest growth rate (1.1), corresponding to the imperial average (1.1), followed by Moesia with no Romance continuation and with a growth rate slightly over the imperial average (1.3), while the top trio consists of Venetia et Histria (1.5), Gallia Narbonensis (2.4), and Lusitania (3.2), in all of which we have a persistent Romance continuation. Here only the position of Moesia would require an explanation – which at this point could only be hypothetical and tentative, so we are not discussing it in detail right now. We can only risk a short explanation: later Moesia might have been the temporary settlement area of the Romanized population that left Roman Dacia after its abandonment and evacuation in the last third of the 3rd century, who were later forced to leave the territory of Moesia as well due to the waves of migration and later the Slavic tribes in the 5th and 6th centuries.[18] This increase in data density in later Moesia might reflect this temporary increase in a Latinized population which later vanished from there.

3.4 Analysis according to the number of linguistic data

Now let us turn to Charts 4a–4c and 5a–5c in Figures 2d and 2e, which exclusively deal with the purely linguistic data, excluding orthographic and extra-linguistic data as non-linguistic ones.

Figure 2d: Charts 4a–4c.
Figure 2d:

Charts 4a–4c.

Figure 2e: Charts 5a–5c.
Figure 2e:

Charts 5a–5c.

In Chart 4a we can immediately notice that in the early period there is again no correlation to be detected between data of the provinces and future Romance developments. In the later period, however (cf. Chart 4b), there is a clear correlation. What is more, both Charts 4a and 4b have a very similar or nearly identical pattern and the same ranking order of provinces as Charts 2a and 2b displaying both the linguistic and non-linguistic data. In both Charts 2b and 4b, in the top positions we find Gallia Narbonensis, Lusitania, and Venetia et Histria; in the middle, Dalmatia; and at the bottom, Moesia, Britannia, and Pannonia. At first glance this could mean that separating linguistic and non-linguistic data was not too profitable, but, as we shall see, such a conclusion is not justified at all. The advantage of such a separation is clearly visible if we go to Chart 4c, displaying the changes in the number of purely linguistic data between the early and the later period, and compare it with Chart 2c, which includes the non-linguistic data as well. Firstly, regarding data growth, the prominence of Venetia et Histria (+264) and Lusitania (+108) as leading terrains of Latinization is much more evident in Chart 4c than in Chart 2c (Venetia et Histria +142, Lusitania −75). Secondly, Dalmatia (−965) with an extinct Romance continuation precedes Moesia (−1,140) with no Romance continuation in Chart 4c, while in Chart 2c Dalmatia (−2,085) lags behind Moesia (−1,686). Through the analysis excluding the non-linguistic phenomena (as in Chart 4c) we were able to identify a more conspicuous concordance with the Romance outcomes than through an analysis including non-linguistic phenomena as well (as in Chart 2c).

3.5 Analysis according to the rates of linguistic data

A further benefit of the method of excluding non-linguistic phenomena can be seen in Charts 5a–5c, displaying the rates of linguistic data as measured against non-linguistic phenomena.

Here not so much the separate percentages by province are interesting and expressive (displayed in Charts 5a and 5b), but their comparison displayed in Chart 5c, where we can see the changes in the rates of linguistic data between the early and the later period. It is of course a welcome sight that all provinces having some kind of Romance continuation are in the top positions: Gallia Narbonensis (+14), Dalmatia (+13 %), Venetia et Histria (+12 %), and Lusitania (+9 %), even if Britannia with no Romance continuation seems to join them (+12 %) as an odd one out (in need of a special explanation). It is even more interesting to find Pannonia in the last position, as the only province where the rate of linguistic data did not increase but instead decreased (−3 %). This means that while linguistic Romanization, i.e. Latinization, the spread of vulgar Latin became more and more intensive and advanced in nearly all selected provinces and also on an imperial average (+8 %), in Pannonia it started to recede and weaken, paving the way to a complete evanescence of any Latin-speaking communities from Pannonia, hindering the potential birth of a Romance language there.[19]

3.6 The inadequacy of a comparative analysis of epigraphic corpora

Before we draw our final conclusions, a short detour seems to be necessary to the question why a simple comparative analysis of epigraphic corpora of selected provinces (by including all inscriptions, that is, also correct inscriptions without any misspellings) is not sufficient for answering our initial questions. Firstly, in terms of chronological classification, we have no reliable figures for the actual number of inscriptions in either the early or the later periods in any of the selected provinces. We only know how many inscriptions survived, but not all surviving inscriptions are properly dated – which would be vital for our analysis. Only inscriptions from a few provinces are fully registered and dated in epigraphic databases; in our case, this means only Pannonia, Dalmatia and Moesia, fully processed in the Epigraphic Database Heidelberg, which we can consider the golden standard. As for Venetia et Histria, we can use the data in Epigraphic Database Roma, but this province is not yet fully processed in the EDR. As for Lusitania, unfortunately we cannot use the Hispania Epigraphica Online database, since its search module does not have a query possibility either for datation or for religion (the latter is vital for differentiating Christian and non-Christian inscriptions). For Gallia Narbonensis, we have nothing but the data of the Epigraphik-Datenbank Clauss / Slaby (EDCS).[20] As a result, what we could do for every province concerned here was to use the most comprehensive database of EDCS, which only allows a modest chronological distinction of inscriptions (as most of the available dates are still waiting to be uploaded into this database) but at least has a query option for Christian and non-Christian inscriptions. Since non-Christian inscriptions can mostly be equated with pre-Christian inscriptions, a division between non-Christian and Christian inscriptions can be roughly equated with a division between inscriptions dated to the early and the later period. Accordingly, Charts 6a–6c in Figure 2f display the distribution of non-Christian (i.e. mostly pre-Christian) and Christian inscriptions, and the changes (i.e. decreases) in the number of inscriptions between the non-Christian (i.e. pre-Christian) and the Christian period as shown in the EDCS.[21] To supplement these data, I also added figures from the EDH for the provinces fully processed there.[22]

Figure 2f: Charts 6a–6c.
Figure 2f:

Charts 6a–6c.

As you can see, however much Chart 6b seems to reflect the future Romance situation, it fails entirely in the leading position, where you can see a high prominence of later Dalmatia with an extinct Romance continuation. This ranking fails to express that the territory of Roman Dalmatia was far from having the most flourishing Romance population. Precaution is also necessary while using this kind of data of the EDCS. While regarding later Dalmatia the figure for Christian inscriptions in EDCS (1,282) and that of inscriptions dated to the late period (i.e. 301–700 AD) in EDH (1,253) are essentially identical (2.3 % difference), the same cannot be said of data concerning later Moesia and Pannonia: Pannonia EDCS 99 (104 by including “sigilla impressa”) vs. EDH 235 (57.9 % difference), and Moesia EDCS 110 (114 by including “sigilla impressa”) vs. EDH 167 (34.1 % difference).

The other reservation about involving all inscriptions regardless of whether they are correct or not, that is, whether they do or do not have misspellings, is rather theoretical. The presence of correct inscriptions attests the presence of the Latin language, but not necessarily the presence of Latin speakers (meaning people who spoke Latin as a mother tongue or as a first or main language after a language shift from the vernacular language). Correct inscriptions may be erected by people who did not speak Latin at all, if they copied formulas.[23] However, inscriptions containing linguistic errors suggest the presence of people who actually spoke the language to some degree. This way only incorrect inscriptions can give positive evidence of the presence of local speakers of Latin.

4 Conclusions

To sum up, we can try and answer the questions asked in the introduction of my paper in a mostly positive way. The process of Latinization of the Empire is traceable by the data of our Database. According to our data, the linguistic Romanization, i.e. Latinization, the spread of spoken or vulgar Latin, became variably more and more intensive in all concerned provinces, since data density increased in the inscriptional sources, even if to different degrees (cf. Chart 3c in Figure 2c). What is more, in many investigational aspects i.e. regarding the patterns of distribution in number of data of all kinds (Chart 2b), of purely linguistic data (Chart 4b), and of rate changes in data density (Chart 3c), we were able to find differences between the selected provinces of the Roman Empire corresponding mostly to the future Romance outcomes of the respective areas (regardless of whether that outcome was positive or negative).

The picture was then successfully refined by an analysis excluding non-linguistic phenomena (4c) that produced a more evident and explicit concordance with Romance outcomes than that of including the non-linguistic phenomena (e.g. Chart 4c). Thanks to this refined method of excluding non-linguistic phenomena from the survey (as used in Charts 4a–4c and 5a–5c of Figures 2d and 2e), we were able to determine (i) the actual position of Dalmatia corresponding better to the Romance outcomes (cf. Chart 4c compared to 2c), and (ii) that in late Pannonia the process of Latinization became slower, since here the rate of linguistic vs. non-linguistic data decreased (by 3 %). This finding helps us better understand the situation in late Pannonia characterized by an explicit decline of Latinization in the 4th–5th centuries, hindering the potential birth of a Romance language there. Accordingly, it is evidently pointless to use such (now obsolete) concepts as “Pannono-Romanisch,” which Reichenkron originally came up with to create a link between the Western Romance Rhaeto-Romance and the Eastern Romance Rumanian language.[24]

All in all, the analysis of data in our Database can contribute to solving the complex problem of Latinization, and is a lot more appropriate for this purpose than a simple comparative analysis of epigraphic corpora of the selected provinces.

Corresponding author: Béla Adamik, Lendület (‘Momentum’) Research Group for Computational Latin Dialectology, ELKH Research Centre for Linguistics, Benczúr utca 33, Budapest, 1068, Hungary; and Department of Latin, Eötvös Loránd University, Múzeum körút 4/F, Budapest, 1088, Hungary, E-mail:

Acknowledgments

The present paper was prepared within the framework of the NKFIH (National Research, Development and Innovation Office) projects no. K 124170 and 135359 entitled “Computerized Historical Linguistic Database of Latin Inscriptions of the Imperial Age” (see: http://lldb.elte.hu/) and of the project “Lendület (‘Momentum’) Research Group for Computational Latin Dialectology” (Research Institute for Linguistics). I wish to express my gratitude to Silvia Tantimonaco for her valuable comments and Zsuzsanna Sarkadi for her help in the revision of the English text.

References

Adamik, Béla. 2003. Die verlorene Romanität in Mösien, Thrakien und Pannonien. In Gerhard Ernst, Martin-Dietrich Gleßgen, Christian Schmitt & Wolfgand Schweickard (eds.), Romanische Sprachgeschichte. Ein internationales Handbuch zur Geschichte der romanischen Sprachen 1, (Handbücher zur Sprach- und Kommunikationswissenschaft 23[1]), 675–683. New York & Berlin: Walter de Gruyter.10.1515/9783110146943.1.6.675Search in Google Scholar

Adamik, Béla. 2006. Offizielles Kommunikationssystem und Romanisierung. In Carmen Arias Abellan (ed.), Latin vulgaire – latin tardif VII.: Actes du VII ème Colloque international sur le latin vulgaire et tardif, Séville, 2–6 septembre 2003, 17–29. Sevilla: Universidad de Sevilla.Search in Google Scholar

Adamik, Béla. 2009. In memoriam József Herman: von der Late Latin Data Base bis zur Computerized Historical Linguistic Data Base of Latin Inscriptions of the Imperial Age. Acta Antiqua Academiae Scientiarum Hungaricae 49. 11–22.10.1556/AAnt.49.2009.1.2Search in Google Scholar

Adamik, Béla. 2012. In search of the regional diversification of Latin: Some methodological considerations in employing the inscriptional evidence. In Frédérique Biville, Marie-Karine Lhommé & Daniel Vallat (eds.), Latin vulgaire – latin tardif IX.: Actes du IXe colloque international sur le latin vulgaire et tardif, Lyon, 6–9 septembre 2009, 123–139. Lyon: Maison de l’Orient e la Méditerranée.Search in Google Scholar

Adamik, Béla. 2015. The periodization of Latin: An old question revisited. In Gerd V. M. Haverling (ed.), Latin linguistics in the early 21st century: Acts of the 16th International Colloquium on Latin Linguistics, Uppsala, June 6th–11th, 2011, 640–652. Uppsala: Uppsala Universitet.Search in Google Scholar

Adamik, Béla. 2016. Computerized Historical Linguistic Database of the Latin Inscriptions of the Imperial Age: Search and charting modules. In Ádám Szabó (ed.), From Polites to Magos: Studia György Németh sexagenario dedicata, 13–27. Budapest & Debrecen: University of Debrecen.Search in Google Scholar

Adams, James N. 2007. The regional diversification of Latin, 200 BC–AD 600. Cambridge: Cambridge University Press.10.1017/CBO9780511482977Search in Google Scholar

Adams, James N. 2013. Social variation and the Latin language. Cambridge: Cambridge University Press.10.1017/CBO9780511843433Search in Google Scholar

Binder, Vera. 2006 [1999]. Latinization. In Hubert Cancik, Helmuth Schneider, Christine F. Salazar, Manfred Landfester & Francis G. Gentry (eds.), Brill’s New Pauly, Antiquity volumes. Brill: Leiden & Boston. https://doi.org/10.1163/1574-9347_bnp_e632280 (accessed 29 January 2019) = Binder, Vera. 1999. Latinisierung. In Hubert Cancik & Helmuth Schneider (eds.), Der Neue Pauly: Enzyklopädie der Antike 6, 1175–1174. Metzler: Stuttgart & Weimar.Search in Google Scholar

Garcia y Bellido, Antonio. 1972. Die Latinisierung Hispaniens. In Hildegard Temporini (ed.), Aufstieg und Niedergang der römischen Welt 1, 1, 462–500. Berlin & New York: De Gruyter.10.1515/9783110818352-017Search in Google Scholar

Herman, József. 2000. Vulgar Latin. translated by R. Wright. University Park, PA: The Pennsylvania State University Press.Search in Google Scholar

Mihăescu, Haralambie. 1993. La romanité dans le Sud-Est de l’Europe. Bucureşti: Editura Academiei Române.Search in Google Scholar

Raupach, Manfred. 1996. Expansion und Rückzug des Lateins. In Günter Holtus, Michael Metzeltin & Christian Schmitt (eds.), Lexikon der Romanistischen Linguistik 2, 1. Latein und Romanisch: Historisch-vergleichende Grammatik der romanischen Sprachen, 5–19. Niemeyer: Tübingen.10.1515/9783110938364.5Search in Google Scholar

Reichenkron, Günter. 1959. Das Ostromanische. In Balduin Saria (ed.), Völker und Kulturen Südosteuropas: Kulturhistorische Beiträge, 153–172. München: Südosteuropa-Verlagsgesellschaft.Search in Google Scholar

Smith, Colin. 1983. Vulgar Latin in Roman Britain: Epigraphic and other evidence. In Wolfgang Haase (ed.), Aufstieg und Niedergang der römischen Welt 2, 29, 2, 893–948. Berlin & New York: De Gruyter.10.1515/9783110847031-007Search in Google Scholar

Spickermann, Wolfgang. 2006 [2001]. Self-Romanization. In Hubert Cancik, Helmuth Schneider, Christine F. Salazar, Manfred Landfester & Francis G. Gentry (eds.), Brill’s New Pauly, Antiquity volumes. Brill: Leiden & Boston. https://doi.org/10.1163/1574-9347_bnp_e1024520 (accessed 29 January 2019) = Spickermann, Wolfgang. 2001. Romanisation. In Hubert Cancik & Helmuth Schneider (eds.), Der Neue Pauly: Enzyklopädie der Antike 10, 1121–1122. Metzler: Stuttgart & Weimar.Search in Google Scholar

Woolf, Greg. 2006 [2001]. Romanization. In Hubert Cancik, Helmuth Schneider, Christine F. Salazar, Manfred Landfester & Francis G. Gentry (eds.), Brill’s New Pauly, Antiquity volumes. Brill: Leiden & Boston. https://doi.org/10.1163/1574-9347_bnp_e1024530 (accessed 29 January 2019) = Woolf, Greg. 2001. Romanisierung. In Hubert Cancik & Helmuth Schneider (eds.), Der Neue Pauly: Enzyklopädie der Antike 10, 1122–1127. Metzler: Stuttgart & Weimar.Search in Google Scholar
Published Online: 2021-06-07
Published in Print: 2021-07-27

© 2021 Béla Adamik, published by De Gruyter, Berlin/Boston

This work is licensed under the Creative Commons Attribution 4.0 International License.

Articles in the same Issue

  1. Frontmatter
  2. Articles
  3. Romanization and Latinization of the Roman Empire in the light of data in the Computerized Historical Linguistic Database of Latin Inscriptions of the Imperial Age
  4. Los patronímicos en los grammatici Latini. La adaptación de una categoría importada
  5. Negative adjectival morphology in Latin.
  6. Does Prenestinian fe⋮faked actually exist?
  7. Ancient Latin grammarians on suppletion
https://doi.org/10.1515/joll-2021-2016
Keywords for this article
database; inscriptions; Latinization; Romanization; vulgar Latin
Creative Commons
BY 4.0

Articles in the same Issue

  1. Frontmatter
  2. Articles
  3. Romanization and Latinization of the Roman Empire in the light of data in the Computerized Historical Linguistic Database of Latin Inscriptions of the Imperial Age
  4. Los patronímicos en los grammatici Latini. La adaptación de una categoría importada
  5. Negative adjectival morphology in Latin.
  6. Does Prenestinian fe⋮faked actually exist?
  7. Ancient Latin grammarians on suppletion
Sign up now to receive a 20% welcome discount
Institutional Access
How does access work?
Have an idea on how to improve our website?
Please write us.
© 2025 De Gruyter Brill
Downloaded on 12.9.2025 from https://www.degruyterbrill.com/document/doi/10.1515/joll-2021-2016/html
Scroll to top button