International Handbook of Language Acquisition

Jessica Horst, Janne von Koss Torkildsen

• Publisher: Routledge
• Year: 2019
• ISBN:
  9781351616614
• Edition: 1
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

How do children acquire language? How does real life language acquisition differ from results found in controlled environments? And how is modern life challenging established theories? Going far beyond laboratory experiments, the International Handbook of Language Acquisition examines a wide range of topics surrounding language development to shed light on how children acquire language in the real world.

The foremost experts in the field cover a variety of issues, from the underlying cognitive processes and role of language input to development of key language dimensions as well as both typical and atypical language development. Horst and Torkildsen balance a theoretical foundation with data acquired from applied settings to offer a truly comprehensive reference book with an international outlook.

The International Handbook of Language Acquisition is essential reading for graduate students and researchers in language acquisition across developmental psychology, developmental neuropsychology, linguistics, early childhood education, and communication disorders.

SECTION 1
Foundations of language acquisition

1
Research on First Language Acquisition

A brief history

Eve V. Clark

The study of language acquisition has a long history, dating as far back as the 7th century BC (Campbell, 2006), initially involving much speculation about how to find out what the original language was. As Campbell pointed out, these were pre-scientific studies, and scientific study of language acquisition only began with the systematic collection of observations, comparing them over time, to document the course of acquisition. With more detailed observations also came attempts to systematize the data, to track changes through different stages, and to see whether all children followed a similar path in learning a first language or languages. In this chapter, I first take up early diary studies, then turn to studies of children’s first sounds in perception and production. Next, I take up the kinds of meanings they first attach to words, and how they build up a vocabulary. I then turn to attempts to write grammars for children at different stages as they combine single words into two-, three-, and four-word utterances. Next, I compare two major approaches to the process of acquisition: nativist approaches and interactive approaches, before considering some of the learning mechanisms that have been proposed. Finally, I raise the need for cross-linguistic data on acquisition and the paths children follow in acquiring different languages before researchers can draw a full picture of what is involved in acquiring a first language. I conclude with a few comments on current trends.

Early days

The first records of children’s early language came in the form of diary studies that also recorded many other aspects of development – motor, moral, and cognitive – alongside language. These diaries were mainly kept by researchers at the institutes of child development founded in the 1800s across Europe (Levelt, 2014). Later on, some diarists focused solely on their children’s emerging language (e.g., Major, 1906, English; Stern & Stern, 1907/1928, German; Pavlovitch, 1920, Serbian/French; Leopold, 1939–1947, English/German; Grégoire, 1947–1949, French). These records contain quite detailed records of early language use, in particular of early sounds, early words, and early sentence structure, and therefore allow for some comparisons across children and languages.

The observational data in early diaries, though, present some shortcomings: the transcriptions of the time were orthographic rather than phonetic; there was little or no context for what children said, so their utterances were often simply listed without any context at all. Yet a number of these studies were quite thorough, and so really constitute the first corpora we have access to. And the data still yield useful insights into some aspects of language development – for example, the kinds of sounds produced early on and the early meanings children assign to words (see, e.g., Jakobson, 1968; Clark, 1973).

Diary studies were largely superseded in the US in the 1900s by simple inventories of the words young children produce early on, and by records of vocabulary size used to assess language development in terms of number of words and average sentence-length at each age (e.g., Templin, 1957). However, since researchers did not focus on sentence structure per se, nor on the functions of different utterance-types, they paid little attention to how children used language, or how their usage changed with development over time.

In the 1960s, largely under the influence of Chomsky’s innovative work in linguistics, research in the US shifted its focus to observations of the grammatical structure in children’s utterances, and began to document the emergence of such structure as children went from one-word utterances to two-word combinations and then added further words and grammatical morphemes to mark such distinctions as person, number, case, and tense. These elements were all tracked as precursors to the acquisition of still more complex grammatical constructions. Researchers also began to do experimental studies of infants’ ability to discriminate sounds from birth on, alongside observations of their early production of sounds in babbling, a precursor to their emerging skill in talking, as their first identifiable words appeared from age 1 onwards. Researchers also began to look at the kinds of meanings young children attached to their early words, and the extent to which these initially diverged from conventional adult meanings.

First sounds

The study of the kinds of sounds infants produced, with their general order of emergence in production, was first considered systematically by Roman Jakobson (1968). He related the buildup of phonological contrasts in production by young children to the dissolution of language in adult aphasias. By drawing on observational records of production from numerous diary studies of different languages, Jakobson could propose a general order of acquisition in terms of contrasts for place and manner of articulation for categories of sounds.

Order of acquisition for contrasts in production:

1. [P-B] stops before [M] nasal
2. [T-D] stops before [N] nasal
3. [A] as first vowel sound

Possible early word-forms that could make use of this initial set of contrasts include:

ba, pa, ma, baba, papa, mama, da, ta, na, dada, tata, nana

More recently, researchers have looked closely at children’s earliest productions of sounds, first in babbling, and then in early word-forms, to see whether babble predicts which sounds children will make use of first in early words. They documented what children could and couldn’t produce at ages 1 and 2, and which sounds appeared easy versus hard for production. For example, children can typically produce stops – [t~d, p~b, k~g] – before fricatives and affricates – [f~v, s~z, tʃ~dʒ] – and before clusters – [st-, sn-, sl-, ʃr-, fl-, str-, br-, bl-]. In early attempts at clusters, they typically omit elements like liquids, [l] and [r], and glides, [j] and [w], as well as [s] before consonants, only adding the missing [s] later on. Researchers also documented the types of sounds that are generally produced late in acquisition, as well as noting individual differences among children (e.g., Ferguson & Farwell, 1975; Ingram, 1989; Labov & Labov, 1978; Smith, 1973; Stoel-Gammon, 2011).

In the early 1960s, researchers also began to make systematic experimental studies of the distinctions among sounds that infants could discriminate from soon after birth. They examined different classes of sounds, focusing on infants’ growing ability to discriminate place and manner of articulation (e.g., /ba/ vs. /ga/, or /ba/ vs. /pa/) as well as their ability to distinguish some intonational contours (e.g., rising vs. falling on the same syllable). This work revealed that, like adults, infants displayed categorical perception at a very early age (see Eimas, Siqueland, Juscyk, & Vigorito, 1971; Jusczyk, 1997; Nazzi, Bertoncini, & Mehler, 1998; Chapter 10, this volume).

The major issue for infants, though, is breaking into the actual speech stream. How do they segment adult speech into units? Essentially, they need to be able to recognize recurring sequences of sounds and treat them as chunks that, later on, they can associate with some consistent meaning. Adult talk with infants and young children should help here since adults talking to young children typically produce rather short utterances, with pauses in between, and repeat them frequently, especially those associated with daily routines like dressing, changing, eating, sleeping, and washing (e.g., Ferrier, 1978). In other words, adults’ child-directed speech helps highlight word and phrase boundaries, at the beginnings and ends of utterances, from early on. For example, adults often use fixed frames as they talk with very young children (e.g., There’s a ___, Look at the ___), another way of signaling the initial boundary for the word that follows. And since that word typically presents new information, it is highlighted with heavier stress (Fernald & Hurtado, 2006; Clark, 2010; Grassman & Tomasello, 2010). By 8 to 9 months old, infants appear sensitive to legal sequences of sounds that begin and end syllables in the language around them, and they will listen longer to legal than to illegal sequences (Friederici & Wessels, 1993; Morgan & Saffran, 1995). In short, they begin to identify recurring sequences of sound from early on, and these sequences, stored in memory, are then available for attaching meanings.

First meanings

A few very early meanings are tied to everyday routines like hi, or bye-bye, plus a wave of the hand, as greetings. In general, though, many early word meanings in young children differ from conventional adult meanings in being over-extended. For example, children produce a word like doggie not only to refer to dogs, but also to cats, squirrels, sheep, and horses. Children rely here mainly on similarities of shape, so the initial meaning attached to doggie could perhaps be characterized as something like “four-legged mammal-shaped entity” (Clark, 1973). Some researchers argued that over-extensions were evidence of failures to discriminate the relevant categories, but others pointed out that over-extensions could simply be the outcome of a communicative strategy, where children opt for the “best word available” in their current repertoire in order to refer to the entity in the current focus of attention. One test of this is to look at how children understand the word dog or doggie when they over-extend it in production. If they interpret it in an adult manner, this would be evidence for a communicative strategy. And indeed, over-extenders choose pictures of dogs over other mammals when asked “show me the dog”. In short, they do not over-extend in comprehension even when they are doing so in production (Thomson & Chapman, 1977; Gelman, Croft, Fu, Clausner, & Gottfried, 1998).

This propensity for over-extending early words has been widely observed across many languages, in diary studies, and it even turns up in the cries of young vervet monkeys (Cheney & Seyfarth, 1988). Researchers have estimated that young children over-extend a third or more of their early words at a stage when they have up to 75 words in production (Rescorla, 1980). But as they add new words in production, they narrow down over-extensions and, with more exposure, begin to adhere more closely to adult conventions on word use (Leopold, 1939–1947; Barrett, 1978).

When children encounter new words, they have to assign some meaning to them. Their initial assignments have been called “fast mapping”, assignments that make some sense in the immediate context where they first hear a new word used. Later, they hear further uses of the same word and can refine and add to whatever initial meaning they assigned. Getting to the conventional adult meaning, though, can take a long time. In assigning meanings, children appear to rely on the same pragmatic principles as adults: they assume that each word has a conventional meaning and that whenever there is a difference in form, there must be some difference in meaning. In short, words contrast in meaning. Over and above that, they assume that participants in a communicative exchange are trying to cooperate (see Clark, 1990, 1993; Grice, 1989).

Children build up their vocabulary quite steadily in production, but comprehension always leads (for adults too). By age 6, children acquiring English are estimated to understand about 14,000 words. From then on, they add about 3,000 more words a year in school, so by adulthood, they are estimated to understand between 50,000 and 100,000 words. Finally, as they add words, they organize them into semantic domains or fields of terms related in meaning (Clark, 1995, 2018b; Rescorla, 1981).

When children lack a word for what they want to talk about, from very early on they fill such gaps with coinages, drawing on known words and affixes to construct terms to convey the relevant meaning in context. For example, as young as age 2, children acquiring English use nouns as a source for new verbs, much as adults do (e.g., it’s flagging, of a flag blowing in the wind; he’s pianoing “playing the piano”), but they don’t always observe all the pragmatic restrictions in this process. They also construct new compound nouns to identify subcategories (a fire-dog “a dog found at the site of a fire”, chimney-smoke vs. car-smoke “exhaust”, a plate-egg vs. a cup-egg “fried” vs. “boiled”); and they form new words through derivation, adding affixes to a known root to convey a new meaning (a reacher “someone who can reach a long way”, a cooker “a cook”, a trumpetist “a trumpeter”, and so on) (Clark, 1982, 1993, 2014). Children typically adopt the most productive options first in constructing new words in their language. They later replace these as needed by the conventional terms. This is because many child coinages are not legitimate from an adult point of view since they stand in for existing words with the very same meanings. Existing words in a language pre-empt such coinages (Clark & Clark, 1979; Clark, 1993).

In summary, children assign and use meanings rather consistently from the start. They observe the same pragmatic principles as adults when they use language. They follow adult usage as a guide to conventional meanings and assume that words that differ in form differ in meaning. But they may take months, or even years, to acquire something like the full conventional meanings of many terms. As a result, they have to rely on whatever partial meanings they have acquired so far. Since these generally overlap to some extent with the intended adult meanings, children can communicate quite effectively, even when they have yet to acquire the ful...

Table of contents

• Cover
• Half Title
• Series Page
• Title
• Copyright
• Dedication
• Contents
• List of contributors
• List of reviewers
• Acknowledgments
• Preface
• SECTION 1 Foundations of language acquisition
• SECTION 2 Dimensions of language learning
• SECTION 3 Individual differences in language acquisition
• SECTION 4 Language acquisition in everyday situations
• Glossary
• Index