12: emergence of syntax

the holophrastic stage

one-word stage

what does production look like in the holophrastic stage

• at the onset, their outputs are single-word utterances

• adults often interpret these single-word utterances as meaning more than the word itself

• ex. the adult replies to the child by answering what they interpreted their utterance as: (ex.”milk” can mean yummy, wants more, etc) → hard to tell underlying syntax

at the holophrastic stage is comprehension or production more advanced

comprehension is more advanced than production

• evidence that children are sensitive to position of heads and complements at the holophrastic stage

head vs complement

• head: determines the category of a phrase (ex. verb)

• complement: a phase that is needed to complete the meaning of the head (complements are often arguments that are needed to complete the meaning of a verb, i.e. direct and indirect objects)

verb phrase

in a vp, the head is v, complement is the object of NP

• This structure shows that English is head-initial (i.e, verb + object (VO))

head directionality parameter

• regarding the order of heads and complements, languages are either:

  • head initial: the head of a phrase precedes its complements; or

  • head-final: the head of the phrase follows its complements

• evidence suggests that children acquire the head directionality parameter early (even at one-word stage)

Hirsh-Pasek and colleagues study on children in holophrastic stage: preferential looking paradigm

• Tested 16- to 19-month-old English-speaking children, who were at the one word stage

• Children were exposed to two scenes (using the preferential looking paradigm):

  • In one scene, Big Bird washed Cookie Monster

  • In the other scene, Cookie Monster washed Big Bird

  • At the same time, they heard a sentence like ‘Big Bird is washing Cookie Monster’

• Results: Children preferred watching the matching scene:
  Even though the sentences are “reversible”, when they hear ‘Big Bird is washing Cookie Monster’, they show that they understand that ‘Cookie Monster’ is the complement of ‘wash’.

what is some extra study evidence using novel verbs

across studies, children were exposed to two scenes with novel verbs

• one scene= duck is gorping bunny

• other scene= bunny gorping duck

results:

• Children looked longer at the correct scene, regardless of whether their language was head-initial or head-final.

• Children’s behaviour must be driven by their knowledge of syntactic structure, given that novel verbs were used. In other words, their understanding of word order is not tied to specific lexical items.

when does the holophrastic stage end

1;6

which specific behaviours indicate the transition from holophrastic to telegraphic stage (2-word)

1. Chained one-word utterances:

   • mommy [pause] door

   • pause= cognitive load

   • usually used to describe smt in their environment

2. repetition of the same word several times

   • ‘up,up,up’

   • used to obtain a specific result (ex. pick me up)

what is production like at the telegraphic stage

• Speech contains mostly lexical/content words;

• Few grammatical morphemes (inflectional morphemes & function words) are used;

• Inflection is absent at higher rates in English (impoverished inflectional system) than in languages like Italian or Hungarian (richer inflectional system)

word order during telegraphic stage

• the two-word combo produced by children largely reflect the word order of the target language

  • they use the correct word order even though words are missing

  • baby cry→ the baby is crying

bootstrapping

using the resources the learner has at hand to solve a problem without explicit help from an external source (like adult)

semantic bootstrapping

The use of meaning (in context) to discover syntactic categories and the order in which they occur.

• assumes that children can pick out what is going on (event) and who is doing what, through experience with the real world: children can thus map semantic categories onto syntactic categories relatively easily

what are Pinker’s findings on semantics

• Semantics helps children identify how objects and events they experience in the real world are mapped to syntax;

• Children first learn some basic semantic properties of words, then they use this semantic information to bootstrap into syntax.

  • ex. child hears: “Mommy throws the ball”

  • from their experience they know that:

    • Mommy: a person (the one doing something (i.e. agent))

    • ball: a thing (the object)

    • throw: an action

  • they figure out that the doer (subject) comes first, then comes the action (verb), and then the thing being acted on (object)

Pinker’s observations and assumptions

• Linguists define lexical categories according to their grammatical function, rather than their meaning:

  • nouns are words that can be preceded by a determiner;

  • verbs are words that can be marked for tense.

• But there are consistent and reliable correspondences between lexical categories (nouns; verbs) and semantic features (objects; events):

  • nouns refer to people, animals and physical objects;

  • verbs refer to actions and changes of state.

an example of semantic bootstrapping

• If children know/have learned the following through experience with the real world...

  • Noun vs. verb: distinguish OBJECTS from EVENTS;

  • Subject vs. object:

    • subjects are typically animate and AGENTS of actions (volitionally causing an event or change of state); and

    • objects are THEMES/PATIENTS (acted-upon entities)

    ...and if they have learned...

  • Meanings for some nouns (like dog, cat, daddy, banana)

  ...then:

  • When they observe a scene where subject, verb and object are consistent with the correspondences identified above (AGENT, EVENT, THEME), they can determine from sentences like:

  1. (1)  The dog is chasing the cat

  2. (2)  Daddy is eating a banana

  that English is head-initial (VO) and that subjects precede predicates (SVO).

what do children do when the semantic cues to lexical categories and sentence structure are unavailable

Structure-dependent distributional learning:

• Children use the rules of the mini grammar they have built to determine the structure of semantically non-transparent sentences.

• The child can use sentences like (1) and (2) to build a mini grammar:

  (1)  The dog chased the cat

  (2)  Daddy is eating a banana

Mini grammar: children learn that

• nouns can be preceded by determiners, like the and a;

• when an event is complete, it is marked by -ed.

• They can extend this grammar to semantically less transparent cases like (3):

(3) The decision evoked a harsh response

• know the category of decision bc of “the”

• they notice that verbs are often marked by past tense

  • Even if children don’t understand harder words like “evoked,” they can understand which word is a noun, verb, etc based on what they know from mini grammar