Language and Cognition Colombo June 2011 Day 1

Language and Cognition Colombo, June 2011 Day 1 Introduction to Linguistic Theory, Part 2

Plan • Identifying syntactic categories • Words and morphemes • Brown’s stages of morphosyntactic development • MLU (mean length of utterance)

Nouns • can have a plural –s inflection • cannot take any of the following inflectional and derivational affixes: present tense –ing comparative –er negating un- or inadverbializing –ly nominalizing –ness • can be premodified by adjectives or a determiner • can complete a sentence like We have no ----- • Semantic information: mass nouns vs. count nouns

Verbs • can have a present tense –ing inflection • some can take an un- or in- negating prefix (e. g. untie, invalidate) • cannot take any of the following inflectional or derivational affixes: plural –s comparative –er adverbializing –ly nominalizing –ness • • can occur after infinitival to can complete a sentence like They / it can -----can be modified by an adverb can be transitive, intransitive or ditransitive

Adjectives • can take the following derivational affixes: comparative –er negating prefixes un- or inadverbializing suffix –ly nominalizing suffix –ness • cannot take the following inflectional affixes: plural –s present tense –ing • can occur after very (if gradable) • can be stacked • can occur in structures like She is very ------

Adverbs • tend to end in –ly • can take the following derivational affixes: comparative –er negating un- or innominalizing –ness • cannot take the following inflectional and derivational affixes: plural –s present tense –ing adverbializing –ly • • can occur after very (if gradable) can appear in structures like She ran very ----cannot appear in structures like She is very ---can postmodify the verb in structures like She behaved ---He treats her ----

Determiners • premodify nouns, and determine the referential or quantificational properties of the noun expressions that follow them • determiners come in two main types, defined semantically: referential determiners – tell us about referential properties of the noun articles : the, a possessives: my, your, his demonstratives: this, that, these, those some other things – e. g. both, either • quantificational determiners – tell us about quantificational properties of the noun some, many, most, every, all

Determiners • determiners cannot be stacked ( so are not like adjectives) • sometimes two determiners can appear together, but such constructions are very restricted and seem to be limited to the co-occurrence of a quantifier and another Det the many books, all my children • determiners (apart from possessives and the) seem to have individual restrictions as to the kinds of noun they can modify – again, this shows that Det are not like Adj: a modifies a singular count noun much modifies a mass noun several modifies a plural count noun more modifies a plural count noun or a mass noun demonstratives agree for number

Pronouns • the most familiar kind is personal pronouns: these do not = people • they encode the grammatical properties of PERSON and NUMBER, and they also vary with regard to GENDER and CASE PERSON NUMBER 1 1 2 3 3 Singular Plural Sing / Pl Singular Plural GENDER M/F M/F Masculine Feminine Neutral M/F CASE NOMINATIVE ACCUSATIVE I we you he she it they me us you him her it them

Pronouns • so pronouns do not pick out an entity in the world, like nouns do; they encode bundles of grammatical features • some words belong to the class of both determiners and pronouns; because most of these do not change their form depending on their use, we have to use substitution to decide which class they belong to in a particular sentence • Some children were ill / Some were ill • Both cats were sick / Both were sick • I prefer this book / I prefer this • I don’t have any cigarettes / I don’t have any • No student failed the test / None failed the test • It’s my teddy / It’s mine

Auxiliaries • can only be followed by another verb • mark grammatical properties of the following verb: tense aspect voice mood • can undergo inversion to form yes/no questions • can be directly negated by a following not • can appear in sentence-final ‘tag questions’

Infinitival to • infinitival to is like an Aux as well: Chomsky compared it to the auxiliary should – they appear in the same positions in a clause – they both take an infinitival verb as a complement – they both allow ellipsis of their complement • infinitival to is not like a preposition: it takes a verb phrase as its complement (P takes a nominal element), and it can’t be modified by right or straight • infinitival to and Aux are both labelled I, or INFL (for INFLECTIONAL ELEMENT)

Complementizers • are grammatical markers that indicate a complement clause The report that war had broken out I don’t know whether she smokes I am anxious for you to do well • they indicate whether the clause they introduce is tensed or not • they mark illocutionary force of the complement clause – i. e. its semantic / pragmatic function: declarative interrogative resultative

Complementizers • not like prepositions: they take a whole clause as a complement (prepositions take a nominal complement) I am hoping for a pay rise =P I am hoping for you to enjoy this class = C • not like determiners: phonological evidence I’m not sure that you did it right I want that book

Prepositions • note that prepositions fall in many ways between the lexical vs functional divide • many prepositions do have antonyms, and therefore could be thought of as having descriptive content (e. g. under/over, to/from, with/without, in/out, up/down) • but many prepositions do not have antonyms, and do not seem to pick out any particular spatial or temporal relationship; they perform a function, such as case assignment (e. g. of, by, for…. . ) • prepositions are a closed class

Prepositions • do not vary their form, so cannot take any of the following inflectional and derivational affixes: plural –s present tense –ing comparative –er negating prefixes un- or inadverbializing –ly nominalizing –ness • can often be intensified by a word like right or straight • can be transitive or intransitive

Substitution • When you’re not sure what kind of phrase or word an item is, use substitution • This just means taking a word or phrase you are sure of, and inserting it in place • If you end up with a grammatical sentence, you know the category of the item you’re working with • If not, try something else (morphosyntactic evidence, semantic or phonological information, educated guesswork…)

Exercise • Identify the word class of each of the italicized words in the following sentences. Give reasons for your analysis and identify any problems. 1. This hedgehog is eating a beetle. 2. The car came round the corner. 3. The moon is round tonight. 4. The exams have come round again. 5. I gave you a book for her. 6. Fred asked if he should explain things to her. 7. He is anxious for her to do well. 8. He must really squirm. 9. Have you got any idea what that was about? 10. Karen said that she hadn’t got any.

Change of pace… • We’re starting to have some background a vocabulary that allows us to discuss problems of language in a relatively technical way • Before we continue to develop those tools, we’re going to lay some foundations for putting all this information into perspective • On to some background/revision about child language acquisition, especially with reference to morphology and syntax

Roger Brown – – – Studied Adam, Eve and Sarah Wrote: “A First Language” (1973) MLU 14 grammatical morphemes Trained ~90% of the top language acquisition researchers • Divided the language development into 5 stages • (mainly focused on I, II)

Brown’s five stages • • • I: 1. 0 -- 2. 0 MLU II: 2. 0 -- 2. 5 MLU III: 2. 5 – 3. 0 MLU IV: 3. 0 – 3. 5 MLU V: 3. 5 – 4. 0 MLU

One-word speech • FIrst utterances are single words – plus occasional “unanalyzed” phrases – Look-at-that Open-the-door • rarely use words simply for naming objects • Often expresses relations and predicates (aboutness) • “Daddy” = Daddy’s slippers • When using language in everyday speech we do not simply name things, but we say things about them.

Semantic Relations in 1 -word stage • POSSESSION: “Daddy” (=slippers) • IMPERATIVE “Open” = open the jar “blow” = blow my nose • NEGATIVE “no” = negate actions • LOCATION “down” = getting down from high chair • RECURRENCE “more” • DISAPPEARANCE “allgone”

Stage 1 • MLU = 1. 75 : First Word Combinations • Productive use of combinations • Lack of inflection – MLU (morphemes) = MLU (Words) • Lack of function words • Assignment: Calculate MLU for one of the Adam files in CLASSWEB

Stage 2 • MLU = 2. 25 : Development of inflections • MLU (morphemes) > MLU (words) • Telegraphic

Telegraphic Speech • Tends to be missing the functors (= closed class words = grammatical morphemes) – Redundant – Have very little semantic content – Children focus on meaning (? ) – only use content words

2 -Word Combinations want daddy want that want here want more Daddy go truck go that up mommy up

Acquisition Order for Grammatical Morphemes (Brown, 1973) Order 1 2 -3 4 5 6 7 8 9 10 11 12 13 14 Morpheme Present Progressive Prepositions Plural Irregular Past Tense Possessive Uncontractible Copula Articles Regular past tense 3 rd person present tense, reg 3 rd preson present tense, irreg Uncontractible auxilliary Contractible copula Contractible auxiliary Example I driving in, on balls broke, fell Daddy's chair This is hot a, the She walked He works She does Ross is winning He's a clown She's drinking

14 Grammatical Morphemes • Development of the closed class - Brown’s 14 morphemes. • 90% obligatory contexts • consistency • Parental frequency - not correlated • semantic & syntactic complexity - yes

90% Obligatory Contexts

Calculating MLU • MLU is based on the average length of a child’s sentences • The length is determined by morphemes rather than by words • What does addition of a morpheme indicate? • MLU calculation procedure: – Transcribe children’s conversation – Divide the conversation into utterances – Divide the utterances into morphemes – Count the number of morphemes in the first 100 utterances, then divide the total by 100

MLU calculations • Speech sample: • “Mommy, want cookie. No dinner! Drink juice. ” • How many utterances? • How many morphemes? • What is the MLU?