The Oxford Handbook of Linguistic Typology

Get Started. It's Free
or sign up with your email address
Rocket clouds
The Oxford Handbook of Linguistic Typology by Mind Map: The Oxford Handbook of Linguistic Typology

1. [Introduction] Setting the stage (Song)

1.1. Gabelentz - morphological typology

1.2. Greenberg - (morpho)syntactic typology

1.2.1. Early rejection of absolute universals as unhelpful

1.3. Universals - what is possible in language

1.3.1. Just like Generativism sought to describe only possible languages

1.3.2. Separation of linguistic tendencies v. non-linguistic influences on language

1.3.2.1. Assuming there's any content to the first category in the first place

2. 1 - The (early) history of linguistic typology (Ramat)

2.1. Classical Antiquity

2.1.1. Some interest in German tribes (Herodotus), but nothing comparative

2.1.2. Thrax's categories were assumed for all other languages

2.1.3. Parts of speech were defined by formal properties, with some semantics (still the same today)

2.1.4. Classical Writers

2.1.4.1. Dionysius Thrax

2.1.4.1.1. Parts of speech

2.1.4.1.2. Grammatike (rhetoric, literary criticism)

2.1.4.2. Varro

2.1.4.2.1. De lingua latina

2.1.4.2.2. Relates that there was an interest in etymology to find the 'true' meaning of words

2.1.4.3. Priscian

2.1.4.3.1. De arte grammatica (most famous grammar of the middle ages)

2.1.4.4. Plato

2.1.4.4.1. Kratylos

2.1.4.5. Aristotle

2.1.4.5.1. Categories

2.1.5. Terms & Concepts

2.1.5.1. grammatike

2.1.5.1.1. rhetoric, literary criticism, parts of speech

2.1.5.2. differentia linguarum

2.1.5.2.1. the difference among languages; early typology

2.2. Middle Ages

2.2.1. Writers

2.2.1.1. Dante Aligheri

2.2.1.1.1. De vulgeri eloquentia - divided languages into oc, oil, and si

2.2.1.2. modistae / grammatica speculativa (speculative grammarians)

2.2.1.2.1. Wilhem of Ockham, Duns Scotus, medieval scholastica

2.2.1.2.2. A logical, universal 'grammar' of thought

2.2.2. Ramat doesn't see these pursuits as properly typological

2.3. Renaissance

2.3.1. Vulgar languages become standardized; awareness of linguistic differences arose

2.3.2. Interest in the Babel question; language classification was political rather than scientific - which language was closest to the pure language of Babel?

2.4. 17th &18th Centuries

2.4.1. Writers

2.4.1.1. Missionary linguists

2.4.1.1.1. Gave Europe insight into languages of the Americas, and further piqued curiosity into the diversity of languages

2.4.1.2. Rationalist Philosophers

2.4.1.2.1. Francis Bacon, Thomas Hobbes, John Locke, Condillac all wrote on language, but not much on typology specifically

2.4.1.2.2. Leibniz

2.4.1.3. Nicolas Beauzee

2.4.1.3.1. Included data from a variety of languages

2.4.1.3.2. Arrive at what is common to languages (a raisonnee; a theory of language)

2.4.1.3.3. In order to do this, data from many languages is necessary

2.4.1.4. Johann Christoph Adelung / Johann Severin Vater

2.4.1.4.1. Adelung collected Lord's Prayer from 500 languages; Vater continued and published this work as Mithridates: Our Father in 500 Languages

2.4.1.5. Franz Bopp

2.4.1.5.1. Conjugation Systems (Sanskrit)

2.4.1.5.2. Father of modern linguistics; first historical comparative approach

2.4.1.5.3. Ultimate focus still on what's universal (philosophical) about language

2.4.1.6. Wilhem von Humboldt

2.4.1.6.1. Bridge between Rationalist Enlightenment philosophy and early 19th Century Romanticism

2.4.1.6.2. genius

2.4.1.6.3. The basic form of all languages must be the same, since all men are predisposed to language

2.4.1.6.4. Interested in the extralinguistic explanations for why languages are the way they are

2.4.2. Early Historical Typology (morphological typology)

2.4.2.1. August von Schlegel

2.4.2.1.1. Beginnings of morphological typology

2.4.2.1.2. Gave us the original 3-way distinction between language types, with inflectional languages subdivided into synthetic and analytic

2.4.2.2. Adam Smith

2.4.2.2.1. Two types of languages: primitive, analytic, simple (original languages) v. compounded, complex (modern).

2.4.2.2.2. Analytic languages develop into synthetic ones (early postulation of grammaticalization, based on Franz Bopp's Conjugation Systems)

2.4.2.3. Humboldt accepted earlier typologies, but considered them as ideal types rather than rigid classifications. Languages didn't fit neatly into one or another.

2.4.2.3.1. Anticipates later typological approaches that treat language as emergent, or as a complex adaptive system (a spontaneous order).

2.4.2.4. Post-Humboldt (Neogrammarians)

2.4.2.4.1. Antoine Meillet - the only valid classification is historical classification

2.4.2.4.2. Historical linguistics became the fad, with typology taking a backseat

2.4.2.4.3. Theorists of the time proposed revisions to the Humboldt typology, positing form v. formless / word v. wordless languages, but didn't really advance the field of typology (imo)

2.4.2.4.4. August Schleicher

2.4.2.4.5. Nikolaus Finck

2.4.2.4.6. Heymann Steinthal (form v. formless)

2.4.2.4.7. Franz Misteli (word v. wordless)

2.5. Georg von der Gabalentz (1840-1893)

2.5.1. Father of modern typology

2.5.2. Sprachenwissenschaft 'typology'

2.5.3. Canonical opposition between isolating and incorporating

2.5.4. Languages historically cycle between the two types

2.5.4.1. Same sentiment is echoed by some modern linguists

2.5.4.2. Because of the cyclicity, it made little sense to view one type as superior to the other. This was the beginning of a modern wertfrei description of language types (no values attached)

2.5.5. Typology aims to be predictive (accd. to Ramat). Gabelentz: If you have one property, you also have these other properties.

2.5.5.1. More properly, typology aims to document correlations; it is only predictive in a non-temporal sense

2.5.5.2. Gabelentz was still pursuing the general 'character' of the language.

2.5.5.3. Aimed to ground these facts in empirical data, via questionnaires.

2.5.6. Hjelmslev: We aim for an exhaustive typology that shows what structures are possible.

2.5.6.1. Again, not in principle different from the aim of the generative paradigm. This doesn't seem to adequately distinguish typology from non-typological approaches.

2.5.6.2. Ramat also doesn't quote Gabelentz in a way that illustrates Gabelentz' commitment to a complete typology

2.6. Conclusion

2.6.1. Typology works by induction; it is interested in the variation of language (and thus what's universal about it)

2.6.1.1. Ramat's positivism shows through here

3. 2 - The pinoeers of linguistic typology: From Gabelentz to Greenberg (Graffi)

3.1. Gabelentz as father of typology

3.1.1. Value-free classification of languages

3.1.2. Distinction between geneaological and typological classification

3.1.3. Confluence of prior interest in word order and morphological typology / 'character' and ethnopsychological classification

3.1.3.1. Humboldt - shift from morphological to syntactic classification

3.1.3.2. Humboldt, Steinthal - morphological typology as 'psychology' (terminology following Muller)

3.1.3.3. H. Weil - word order (analogical v. transpositive); separated the syntax from the order of thoughts, thus removing (most) valuations of language from the classification (languages that could do both types of word order were better)

3.1.4. A number of his near-contemporaries focused on word order issues, even beginning the practice of using abbreviations like SVO (Wundt), and a typological sample (Schmidt)

3.2. Structuralism

3.2.1. Lost interest in typology, because typology had been linked with ethnopsychology and evaluations, and Saussure and Sapir had established linguistics as a value-free science

3.2.2. More interested in finding differences across languages than their universals

3.2.3. Sapir's book Language shows an awareness of the Schlegels and Steinthal (not mentioned by name), but not Gabelentz (Sapir uses the word 'classification' instead of 'typology'.) Major influence on linguists and their approach to typology.

3.2.4. Sapir establishes a multidimensional typology of morpheme types, with ideal types that no language perfectly realizes

3.2.5. Word order studies developed separately from typology at the time (Tesniere)

3.2.6. Implicational typologies / universals - developed originally by Jakobson for acquisition of sounds, and developed by Hjelmslev in his program for typology

3.3. Greenberg

3.3.1. Focus on the unity in diversity, rather than the diversity itself

3.3.2. 3 types of universals: absolute, statistical, and implicational

3.3.3. Word order was a traditional, easily-studiable empirical topic

3.3.4. Correlates a variety of types of word orders to each other

3.3.5. Greenberg's word order universals became an explanandum for generative linguistics (and still are). Others took a typological approach (distinguishing therefore between formal and typological syntax, still relevant today).

4. 3 - Linguistic typology and the study of language (Daniel)

4.1. How typologists approach the study of language

4.1.1. Interested in diversity as such, which is what makes typology unique as a discipline

4.1.1.1. This seems false. Typologists are ultimately interested in universals of language, just as much as the generativists. This isn't a distinguishing feature of typology, just the typological method.

4.1.2. Contrasts to generativism, which focuses on what is universal (hence Universal Grammar)

4.1.3. Daniel acknowledges that the generative research agenda sounds typological in nature, but notes two methological differences

4.1.3.1. Generativism typically focuses on in-depth data from a small set of languages (a methdologically sound move if you accept the premises of generativism)

4.1.3.2. Generativism is holistic, with researchers all contributing to adjustments to the same theory and paradigm (typology, by contrast, is extremely multi-faceted)

4.1.3.3. Daniel characterizes the main difference between generativism and typology as deductive v. inductive ('generativists *know* that all languages are identical, while typologists seek to ascertain this...' (p. 49))

4.1.4. Modern typologists focus too much on reference (a reaction to the overly relational focus of radical structuralism / generativism). Daniel argues that typology needs to acknowledge both referential (semantic) and relational (structural, formal) explanations for language.

4.1.4.1. Grammaticalization and diachronic studies focus purely on the referential, hence their popularity in modern typology

4.2. The incompatability paradox [i.e. the problem of crosslinguistic categorization]

4.2.1. Saussure's solution was to focus on minimal contrasts, giving birth to Structuralism

4.2.1.1. Saussure himself was reacting to the Neogrammarians [Daniel doesn't specify which aspect of their paradigm, precisely; presumably they focused on the referential / historical meaning of terms, and purely historical explanations for synchronic patterns]

4.3. Typological observations

4.3.1. Taxonomies, scales, parameters, and implicational hierarchies

4.3.2. Gradience in categorization / classification leads typologists to use unidimensional or multidimensional scales

4.3.3. Also important to consider the variation that occurs within a language on the basis of sociolinguistics and pragmatics, etc. (Lahiri & Plank 2008)

4.3.4. Historical typology / typology of language change - how the linguistic system may develop over time, and the trends we see or do not see

4.4. My blog post about this chapter: http://dlc.hypotheses.org/299

5. 4 - Explaining language universals (Moravcsik)

5.1. Types of universals

5.1.1. Unrestricted

5.1.1.1. Absolute: X holds for all languages

5.1.1.2. Probabalistic: X holds for most languages

5.1.2. Restricted [implicational]

5.1.2.1. Absolute: In all languages, X → Y

5.1.2.2. Probabalistic: In most languages, X → Y

5.2. Universals as explanations (explanans)

5.2.1. 3 types of explanations

5.2.1.1. 1. Structural explanations

5.2.1.1.1. These often aren't causal (e.g. word order correlations that explain the order of adpositions in a language still don't explain why such a correlation exists)

5.2.1.1.2. Language acquisition is another type of explanation, somewhat more causal. But it also just pushes the explanatory chain back one level - why did they acquire that grammar rather than some other one?

5.2.1.2. 2. Historical explanations

5.2.1.2.1. These changes also push back the explanatory chain a level - why did a specific language, or all languages, change the way it did?

5.2.1.3. 3. Functional explanations

5.2.1.3.1. Bybee: "Ultimately, we are brought back to the synchronic plane where we must ask what cognitive processes are behind [the new development.]"

5.3. Universals as things that need to be explained (explananda)

5.3.1. Structural - sometimes structural universals can be explained by other, broader or more general structural universals

5.3.2. Historical

5.3.2.1. Universal constraints on change (e.g. pronouns lag behind nouns in historical changes)

5.3.2.2. Universal sources of change (e.g. functional elements derive from lexical elements)

5.3.3. Functional - human cognition and physiology

5.3.3.1. Language use (performance)

5.3.3.2. developmental change (acquisition)

5.3.3.3. historical change (how the language is modified in the process of acquisition)

5.3.3.4. Typically cast as standing in opposition to the innateness hypothesis, but innateness makes reference to cognitive principles as well

5.3.3.4.1. The critical issues is the domain-specificity of functional explanations

5.3.3.5. Hawkins: "grammatical conventions are "frozen" processing preferences"

5.3.3.6. Moravcsik says that functional explanations can only explain language change, not language synchrony

6. 5 - The problem of cross-linguistic identification (Stassen)

6.1. The problem: In order to do a crosslinguistic typology, one has to be able to identify the phenomenon under study

6.1.1. Criteria for typological comparison

6.1.1.1. 1. Language-independent / external definition

6.1.1.2. 2. Researcher must be able to identify both what is, and what is not, an instance of the phenomenon

6.2. No criteria can be entirely formal, but they can be entirely 'semantic'

6.2.1. I'm not 100% convinced of this, but I do definitely agree that nobody has succeeded at this task yet. A purely formal typology would probably look extremely different from typologies that are currently seen as 'formal'.

6.2.2. Croft points out that we should talk of 'external' criteria, rather than 'semantic' criteria. This would allow typologies to encompass criteria based on pragmatics, phonetics, and logical relations.

6.2.2.1. Also comports very nicely with the major insight behind functionalism

6.3. [According to Stassen] External criteria are necessary but not sufficient for typological comparison, at least in practice

6.3.1. Authors almost always adopt mixed definitions, which helps keep the domain of investigation manageable. It's usually possible to divide the phenomenon into distinct external and formal components.

6.4. For Stassen, this is a methodological problem, and not a theoretical one. As such, this chapter is perhaps best interpreted as a guide to thinking about how to frame one's definition for doing a typology.