Language Classification: History and Method

Free download. Book file PDF easily for everyone and every device. You can download and read online Language Classification: History and Method file PDF Book only if you are registered here. And also you can download or read online all Book PDF file that related with Language Classification: History and Method book. Happy reading Language Classification: History and Method Bookeveryone. Download file Free Book PDF Language Classification: History and Method at Complete PDF Library. This Book have some digital formats such us :paperbook, ebook, kindle, epub, fb2 and another formats. Here is The CompletePDF Book Library. It's free to register here to get Book file PDF Language Classification: History and Method Pocket Guide.

Description In the recent past, especially non-Africanist linguists have raised concerns about the reliability of the widely accepted Greenberg classification for African languages and the empirical evidence which underlie it, in particular when compared to standards applied in other philological disciplines and historical linguistics in general.

This small international workshop aimed at discussing problematic parts of Greenbergs classification in the light of new research results. The discussion recognized commonly accepted standards for the establishment of genealogical relationships on all levels and the procedure in historical-comparative linguistics as laid out in such works as Nichols , Campbell , and Campbell and Poser The workshop assembled leading specialists who have been carrying out work relevant for this domain and reviewed their results with respect to still controversial genealogical affiliations in Africa.

A lot of previous classificatory work has focused on lexical comparisons and, though to lesser extent, the accompanying establishment of regular sound correspondences. Since particularly Nichols strongly argues in favor of evidence from grammatical features, particularly of a paradigmatic nature, the workshop focused on this kind of data as evidence for or against a particular genealogical relationship.

DE EN. Events Information for Data Privacy Statement. It is a normal situation, when sound changes are modest and gradual during centuries and then get abrupt and massive. These conservative languages contrast, e. An instance of substantially different rate of phonetic changes within the same clade is the Italic language group.

  • Another Life Altogether!
  • Language Classification History Method by Lyle Campbell - AbeBooks.
  • Human Learning: Biology, Brain, and Neuroscience.

Despite close relationship, French and Italian vary in respect of historical sound changes: Italian is relatively conservative, whereas French is rather innovative. It is illustrated in Table 3 which contains ten French-Italian lexical pairs from the beginning of the Swadesh wordlist accompanied with their Latin protoforms.

In other words, phonetic changes in natural language evolution represent a stochastic process which can hardly be described by evolutionary models.

See a Problem?

As concerns the specific method of consonant classes adopted in the present paper, it should be noted that in language history, the bulk of inferred phonetic shifts is typologically trivial by definition , i. But certainly mutations between different consonant classes are also rather common and almost inevitable.

The ratio of trivial mutations to non-trivial ones is random for individual cases. If an algorithm of marking cognation is based on phonetic similarity, some etymologically related words are marked as non-cognate, whereas some phonetically similar, but etymologically unrelated words are marked as cognate. As follows from the above, phonetic similarity-based cognation marking adds noise into the input matrix as compared with the etymology-based approach.

Since Lezgian languages generally demonstrate a lot of non-trivial phonetic changes, it is interesting how various methods cope with phylogeny reconstruction based on the noisy dataset. Below their results are compared with the consensus etymology-based tree Fig.

The following Lezgian trees with phonetic similarity-based cognations were obtained: Fig. The StarlingNJ tree Fig. In sum, the main flaws of the distance-based methods as compared with the consensus etymology-based tree are the following ones. Thus, for the phonetic similarity-based matrix, I consider the results of the distance-based methods as good.

On the contrary, the character-based methods appear to be less reliable. The main flaws of the UMP tree Fig. As noted above, the Lezgian item database [ 3 ] has several important features. In the theoretical paper [ 50 ], the adequacy of the main phylogenetic methods is tested by simulation of various linguistic situations.

  • Joseph Greenberg.
  • Joseph Greenberg - Wikipedia.
  • German Histories in the Age of Reformations, 1400–1650!

There are, however, specific difficulties in application of the MP method with binary characters to linguistic data. One of the reasons for that can be the unconformity of the used model to our ideas of natural language evolution. The MP method depends on homoplasy i. Correspondingly, for minimization of the effect of homoplastic disturbances, the authors of [ 50 ] propose to use individual costs of characters Weighted maximum parsimony , assigning higher costs to those characters which do not demonstrate homoplasy in the given linguistic data.

Language Classification: History and Method by Lyle Campbell

Thereby to detect linguistic homoplasy, it is needed to reconstruct ancestral character states that is actually a non-trivial theoretical and practical task [ 52 ], particularly the reconstruction is impossible without the established phylogenetic tree—as a result we get in a vicious circle.

As a quantitative assessment, it is proposed in [ 50 ] that all the tested phylogenetic methods, except for UPGMA, reconstruct ca. For the etymology-based input matrix, experiments with the Lezgian lexicostatistical database present, however, a more comforting picture, if one believes that each branch of the true tree has been reconstructed at least by one of the tested methods except for UMP i.

Under the assumption of a relatively small temporal error with the joining of neighboring nodes within such a time span, see Fig. The consensus etymology-based tree Fig. An unexpected result of the Lezgian test is the relatively low plausibility of the obtained UMP-tree Fig. It is a somewhat unexpected result.

Language classification

More tests of this kind are needed to clarify the situation. The examined Lezgian data support some propositions which serve as an ideological basis of the Global Lexicostatistical Database project. I express my thanks to Valery Zaporozhchenko Moscow and Johann-Mattis List Marburg for their valuable advice on phylogenetic software, and to Dmitry Leshchiner Moscow for help with specific mathematical issues.

I would like to thank two anonymous reviewers for their helpful comments and fruitful discussions. In addition, I must note that the present study could hardly be possible without discussions on related or wider topics with George Starostin, Mikhail Zhivlov, Anna Dybo, Philip Minlos and my other colleagues from the Moscow school of comparative linguistics. I am also grateful to Evgeny Satanovsky for his generous support of the Tower of Babel and Global Lexicostatistical Database linguistic projects which has enabled us to conduct valuable research in the field of linguistic phylogenetics for the past several years.

The author remains responsible for all possible errors of fact or interpretation. Conceived and designed the experiments: AK. Performed the experiments: AK. Analyzed the data: AK. Wrote the paper: AK. Browse Subject Areas?

Account Options

Click through the PLOS taxonomy to find articles in your field. Abstract A lexicostatistical classification is proposed for 20 languages and dialects of the Lezgian group of the North Caucasian family, based on meticulously compiled item wordlists, published as part of the Global Lexicostatistical Database project. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited Data Availability: All relevant data are within the paper and its Supporting Information files.

Data Lezgian is a relatively deep linguistic group deeper than the German, Slavic or Turkic groups, but younger than the Indo-European family which consists of languages spoken in South-East Dagestan Russian Federation and the adjacent parts of Azerbaijan, see Fig. Download: PPT. Fig 1. Map of the modern Lezgian lects adapted from [ 1 ]. Methods Lexicostatistical trees were produced by several phylogenetic methods. Modified neighbor joining method, designed by S. Starostin for lexicostatistical analysis and implemented in the Starling software method Starling neighbor joining, hence StarlingNJ ; see [ 21 ].

The distance between two lects A and B is 1 minus the percentage of shared Swadesh items, e. If a non-modern lect is involved, its percentage is automatically adapted to AD according to the accepted molecular clock model [ 21 ], [ 22 ]. Standard neighbor joining method hence NJ , see [ 29 ], [ 30 ]. The trees were produced in the SplitsTree4 software v.

Total characters proto-roots for etymology-based calculations and characters for phonetic similarity-based calculations. The trees were rooted by the outgroup the Chechen wordlist. The trees are not dated. The trees were visualized in the FigTree software v. Also additional trees were produced by the BioNJ method [ 32 ], these appeared to be topologically identical to the NJ ones.

Report copyright / DMCA form for Language Classification: History and Method

Markov chain Monte Carlo method under Bayesian framework hence Bayesian MCMC , see [ 30 ], as it was for the first time applied to linguistic data in [ 34 ]. The trees were produced in the MrBayes software v.

Søren Wichmann

The program was run 4 times using 4 concurrent Markov chains; the Chechen language was marked as an outgroup.