LINGÜÍSTICA COMPARADA

Amaya Mendikoetxea
[email protected]
204-VI BIS

TOPIC 1

PART 1: Typology and universals: Romance and Germanic languages.

There is no consensus on the ultimate origin or age of human language. However,
there are 2 theories about how languages were created: continuity theory and
discontinuity theory.
+ Continuity-based theories stress that language is so complex that it must have
evolved from earlier pre-linguistic systems among pre-humans.
+ Discontinuity-based theories stress that language is a unique human trait that
appeared fairly suddenly in the transition from pre-hominids to early man.

Ideas in the Tower of Babel:

1. Having one language makes you powerful. It is often assumed that language
was given to us as a biological advantage over other species.
2. If you get everybody speaking the same language they’ll work together and in a
more productive way.
There are between 6,000 and 7,000 different languages around the world.
The area of linguistics that deals with this variety of languages is language typology
(area which deals with all sorts of things related to languages).

1.1 LANGUAGE FROM A TYPOLOGICAL PERSPECTIVE.

(3) Explaining the underlying unity of language:

+ Genetic / Biological Approach:
Humans are genetically endowed with a ‘language faculty’, which is specific for
language, distinct from other cognitive capacity. There’s a module in our brain that
deals specifically with language. This language faculty, this module, is essentially a
Universal Grammar (UG) that we all share. This language faculty is also a language
acquisition device (LAD), meaning that if a child is in China, he’s exposed to Chinese
language by several factors such as his family or television programs so he has to build
a grammar for this language, which is filtered by its language faculty. We’re genetically
programmed to create a language so that we can make sense to the sounds we’re
exposed to.

+ Language Use / Functional Approach:

There’s no module of the brain that deals with language. Language comes from
outside and all languages are used for similar purposes (functions): asking questions,
negating facts, making comparisons, commands, etc. speakers develop grammars
which are highly efficient in carrying these functions out. Under the pressure of the s
same communicative tasks, languages evolve such that they exhibit grammatical
similarities.
1.2. WHAT IS TYPOLOGY?

(4) in the old historical linguistics the emphasis was to classify languages. Nevertheless,
the idea of typology is not only to classify languages but also their components.
Typology is ‘the classification of languages or components of languages based on
shared formal characteristics’. According to Whaley, there are 3 propositions which are
part of what typology is:
1. Typology involves cross-linguistic comparison.
2. A typological approach involves classification of either (a) components of languages
or (b) languages.
3. Typology is concerned with classification based on formal features of language
which encode semantic categories.

Proposition 1: Typology involves cross-linguistic comparison

(5) They’re all relative clauses. You have to look the things that they share but also
their differences, if you only looked English and Spanish, you won’t be able to classify
all languages in the world; because of that, we need the Basque example.
They’re all Restrictive Relative Clauses (RRC) where the reference of the noun is
restricted.
(7) In Basque, the order of the words is reversed; the relative clause precedes the
noun. Whereas English and Spanish are head-first languages, Basque is a head-last
language. In head-first languages, if there is an NP, anything that modifies it must go
after it. However, in head-last languages any modifier precedes the noun.

When you do cross-linguistic comparison you’re not only interested in similarities but
also differences. Most differences between (4) and (5) have to do with the element
that introduces the relative clause. In English you can omit the wh-phrase or the
complemetizer using a covert element but that would make a Spanish sentence
ungrammatical.
In English, when the relative element is the complement of a preposition then the
preposition can leave the clause it belongs to but in Spanish the relative marker needs
to have the preposition with it, the preposition can never leave the clause it belongs
to. E.g.: (5a) I have seen the man who you danced with.
(5a) He visto al hombre con quien bailaste.
This is called preposition stranding, something that all Germanic languages have. It’s
something that manifest in relative clauses but it also happens in other structures such
as interrogatives. E.g.: Who did the man dance with?
¿Con quién bailó el hombre?

Proposition 2: A typological approach involves classification of either (a) components

of languages or (b) languages.
We look at particular constructions and see how they are interpreted in different
languages, to get a classification.
Reflexives
(8) a. John knows himself well.
It’s a reflexive construction with an anaphor in an object position.
x knows y x = y x and y refers to the same entity (John), they’re anaphors (a
word or phrase that refers back to an earlier word or phrase).

*He1 knows him1

This sentence is ungrammatical because ‘him’ cannot be used to refer to the same
entity as ‘he’.

*He1 knows himself2

This sentence is ungrammatical because ‘himself’ cannot refer to a different entity
than ‘he’.

Binding theory: (buscar en apuntes de grammar II)

In English reflexive constructions are expressed in means of anaphors that can only
appear in certain positions.
*Himself1 knows John1

b. Juan se conoce bien (a sí mismo).

In Spanish the anaphor is not compulsory. Anaphors are not the only means by which
we can express reflexives.
‘Se’ is not an optional element, it is a pronoun that, unlike ‘himself’, is not in the object
position. There can be nothing intervening between the element ‘se’ and the verb. ‘Se’
is what we call a nominal clitic, an element that attaches to the verb, it’s not an
independent element. We have a clitic with an optional anaphor.
Juan1 le2 conoce. Juan1 se 1 conoce. *Juan se y le conoce.
Juan sólo piensa en sí mismo reflexive using an anaphor. The syntactic context is
different, we have a prepositional verb.

c. Jonek ondo ezagutzen du bere burua.

Jon-ERG well know AUX-3SG his head-ABS
In this case, the reflexive is expressed with a Noun Phrase which expresses that this
verb is reflexive. There are languages which express reflexives by affixes or that don’t
use an explicit marker for the reflexive. E.g. John washed.

Impersonals
(9) Quand on est heureaux, on dot bien. (Fr)
‘When one is happy, one sleeps well.’
The sentence has a generic subject, not a specific one. This generic subject is expressed
by ‘on’ pronoun, equivalent to ‘one’ in English or ‘uno/a’ in Spanish.

(10) Täällä er saa polttaa. (Finnish)

Here not may smoke
‘One can’t smoke here.’
There is no subject, the verb is in 3rd person singular, what indicates that it’s an
impersonal structure.
(11) a. Aqui não pode nadar. (Brazilian Portuguese)
Here not can swim.
Brazilian Portuguese is like Finnish (in terms of impersonality), we don’t have any
marker to express impersonality.

b. Aqui nã se pode nadar.

Here not se can swim.
‘One can’t swim here.’
European Portuguese is like Spanish (in terms of impersonality), ‘se’ expresses
impersonality.

There are different ways to create impersonal structures:

 Use of a pronominal (on, one, uno/a)
 Absence of subject (3rd person singular/plural). E.g. Aquí fuman mucho it’s
ambiguous since it could be a generic or specific statement. ‘Here they smoke a
lot’ ‘they’ can also be interpreted as impersonal.
 Person pronouns like ‘they’ or ‘you’.
 2nd person. E.g. Puedes comprar un montón de cosas aquí / You can buy lots of
things here.
 Clitic or verb affixes: elements that attach to the verb.
There are also other ways to express impersonality, as passives.

→ Typological analysis of components of languages

 Using crosslinguistic data (looking at different languages) all the types of a
specific phenomenon are determined.
 The goal is to better comprehend how this aspect of language operates by
identifying the degrees of similarity and variation one finds among languages.
 The aim is not so much to classify languages but to understand some basic
facts about phonology, morphology, syntax, semantics or discourse.
(Semantics always take a part in typological analysis)
→ Typological analysis of languages
 Aim: classifying entire languages into categories based on shared formal
features (e.g. SVO vs. SOV languages).

Proposition 3: Typology is concerned with classification based on formal features of

language which encode semantic categories.
 Typologists classify languages in terms of the forms out of which a language is
composed: sounds, morphemes, syntax, or discourse structure.
 There is a strong association between typological and genetic classification.
Languages which are related tend to share typological aspects but that’s not
always the case.
 There may also be a strong association between typological and geographical
classification:
o Grammars often adopt some features of other languages that are close to
them spatially, even if they are not genetically related.
o This is particularly common in speech communities where two or more
languages coexist.
PART 2: Comparative syntax: Principles and Parameters
 The comparative approach in linguistics goes back to the 19th century: it had a
historical goal, that of establishing relations and kinships among languages.
 Comparative linguistics (originally comparative philology) is a branch of
historical linguistics that is concerned with comparing languages to establish
their historical relatedness.
 Genetic relatedness implies a common origin or proto-language, and
comparative linguistics aims to construct language families, to reconstruct
proto-languages and specify the changes that have resulted in the documented
languages. To maintain a clear distinction between attested and reconstructed
forms, comparative linguists prefix an asterisk to any form that is not found in
surviving texts. A number of methods for carrying out language classification
have been developed, ranging from simple inspection to computerised
hypothesis testing. Such methods have gone through a long process of
development.

 Since the early 1970s, there has been a return to a comparative approach in
language study. The goal of this ‘new’ comparative approach is ‘psychological’
(not historical):
o We want to know the properties of Universal Grammar, so we need to
study what is constant and what varies among languages
o We want to know what constitutes knowledge of a language X (X = Spanish,
English, Basque…), so we need to know what properties are specific to X
and which are universal, common to all languages.
The idea is to know what kind of cognitive system do we have in the brain that is called
English.

(12) Components of UG:

Principles: rigidly-fixed, predetermined (innate, part of our genetic endowment, our
cognitive capacity to learn a language) (quote Radford 193: 13)
Parameters: choices left open by UG, which are set in different ways by different
languages – subject to cross-linguistic variation. E.g. Null Subject Parameter: in Spanish
you can say ‘vino’, but in English you cannot say ‘came’, you have to say ‘he/she came’

- The language faculty must be such as to allow the child to develop the grammar of
any language given experience (input) of that language
- By positing a set of innate UG principles which don´t have to be learned we are
minimizing the learning burden (i.e. accounting for the rapidity and success of L1
acquisition).

(13) UG as a Language Acquisition Device.

Experience Universal Grammar

of L1 Grammar of L1
Not all aspects of grammar are determined by innate grammatical principles; if it was
so all languages would have the same grammar and there would be no structural
learning in language acquisition, just lexical learning.
 Structural learning is limited to parameters or parametrical aspects of
structure: those aspects which are subject to cross-linguistic variation.

Our brains are built in such a way that we are genetically predisposed to learn a
language. Initially, there are no differences in our brain. Once we learn certain aspects
of the language we’re exposed to, things change and it’s more difficult to learn a new
language. Input experience is essential but it has to be interpreted and we do it
through UG, which works as a language acquisition device.
We have lexical learning and some structural learning (parameters).

(14) Null Subject Parameter.

a. Vinieron ayer/ Llueve/ Es probable que nieve.
Vinieron: 3rd person singular, we could say ‘Ellos vinieron’
Llueve: expletive / dummy subject, you cannot say ‘Ello llueve’
Es probable que nieve: expletive / dummy subject, you cannot say ‘Ello nieve’.
There’s no pronoun in Spanish that you could use as a subject of ‘llueve’.
b. *Came yesterday/ *Rains/*Is possible that it will snow

[ Note: cf. binary setting – Radford is WRONG!!]

(i) Pedro ganhou na loto vs. * ganhou na loto it’s ungrammatical when there’s no an
explicit subject. If the subject is explicit we need a pronoun.
There are partial Null Subject languages. For some people you could have null subject
or not. E.g. ‘vine’ vs. ‘yo vine’. It’s possible in 1st person singular and 2nd person. (mirar
apuntes en sucio)

(ii) a. Pedro disse que Ø ganhou na loto. (Brazilian Portuguese)

Pedro said that won on the-lottery
‘Pedro said that he won on the lottery.’
The previous ungrammatical sentence has been embedded in a largest structure
where it’s grammatical. However, the null subject Spanish sentence is ambiguous. E.g.
‘Pedro dice que ganó la lotería’, we don’t know who is the winner of the lottery.
In Brazilian Portuguese, the null subject has to refer only to ‘Pedro’, if someone else
won the lottery (ii.a), it would be ungrammatical.
b. Aqui não pode nadir.
here not can swim
‘One can’t swim here.’

Parameters as clustering properties:

 Spanish allows an empty pronominal (pro) in subject position
 Spanish allows inversion of the subject (Han perdido su trabajo muchas
personas)
 Spanish lacks overt expletives (*ello llueve/*ello es possible que…)
 Spanish shows no that-trace effects (¿Quién cree María que t ha llamado?/
*Who does Mary think that has called?