First Language Acquisition. How Englisch-speaking children acquire past tense structures

Term Paper (Advanced seminar), 2020

13 Pages, Grade: 1,3


Table of contents

1. Introduction: First language acquisition

2. On learning verbs

3. Acquisition of past tense structures
3.1 Tense and aspect
3.2 Regular and irregular verbs

4. Conclusion

5. Bibliography

1. Introduction: First language acquisition

Language—and especially language acquisition—seems to be the closest thing to magic. Appropriately enough, in his pioneering study A First Language, Brown (1973: 63) poses the rhetorical question whether “the first words [a]re [...] a kind of magic”. Linguists have long been intrigued with children’s acquisition of their native tongue. But only since the 1970s, considerable attention has been paid to first language acquisition in research (Kuczaj & Barrett 1986: ix). First language acquisition is the study of when and how infants and children get a command of their native tongue (Goodluck 1991: 1). Even though there are a number of empirical studies and data, there is still a significant need for further research on children’s language acquisition. The fact that children acquire implicit and productive knowledge of adult grammar—even though they do not obtain explicit instruction in the linguistic rules of their specific language and their language input is severely restricted to the speech that they hear—is called the logical problem of language acquisition (Goodluck 1991: 3). According to Goodluck, this logical problem is the reason why the notion of an innate and unconscious linguistic knowledge is quite common among (psycho)linguists (1991: 3). They claim, Language acquisition can only be accounted for if we assume that children work with knowledge of principles of grammar. The linguistic system involves rules too abstract and complex to be learnt without the aid of innate knowledge about the nature of the system. The general idea is that the child is equipped with a set of blueprints that define and limit what a human language can be like. This innate knowledge goes under the name of universal grammar [...] [which] was influentially laid out and discussed by Chomsky. (Goodluck 1991: 3)

The assumption that the child is biologically equipped with fundamental linguistic knowledge can additionally be justified with the fact that deaf children babble (Goodluck 1991: 141). Various linguists have observed that children seem to undergo three phases of language acquisition (Kuczaj & Barrett 1986: ix-x): The first phase begins approximately around 10-12 months of age with children producing idiosyncratic sounds and adding only one to three words per month to their lexicon. The second phase begins around 16-18 months of age when children start to use words in a referential manner. Vocabulary growth is much quicker now (20-50 new words per month). The third phase begins at some point during the third year of life with children learning 50 or more words per month. English-speaking children aged two to two and a half years old have generally acquired vocabularies consisting of 500 words or more (Barrett 1995: 392). Clark surveys that henceforth, English-speaking children learn approximately ten words a day to reach a vocabulary containing 14,000 words by age six, and from then on, they add at least 3,000 words a year until they are seventeen (1995: 393).The study of first language acquisition is as complex as the process itself since there are different theories and approaches and, most importantly, because language acquisition differs cross-linguistically and individually. Therefore, the following merely attempts to provide an overview of how English-speaking children start to acquire past tense structures. This is particularly interesting because children’s acquisition of relational terms like verbs have only recently been studied in greater detail (Behrens 2001: 451).

2. On learning verbs

“Possibly no other area of the lexicon is as complex as verbs. They not only convey information about events and states but also about motives, intentions, [and] causality” (McShane, Whittaker & Dockrell 1986: 275). Due to the semantic and syntactic complexity of verbs, Gentner argues that nouns—especially concrete nouns—are acquired before verbs (2006: 545-546). Gentner uses the Natural Partitions Hypothesis to explain that verb meanings are relational, variable, and linguistically shaped and therefore not as transparent as referential noun meanings for children (2006: 544-545). This early noun bias in first language acquisition was observed in several studies in developmental linguistics (Tomasello 1992: 210). For example, in 2002, Childers and Tomasello taught 2-year-olds novel verbs and nouns and observed that the English-speaking infants acquired nouns at a much faster rate than verbs (Gentner 2006: 547). Furthermore, Tomasello analysed his daughter’s early speech development in a diary and noticed that she learned concrete nouns before relational terms such as verbs (1992: 210). English as a language uses morphological devices to modulate the meanings of words. Consequently, a child learning the verb draw will encounter the forms draw, draws, drew, drawn, and drawing. Hence, the difficulty in learning verb meanings in English is attributable to the fact that the language-learning child has to trace the identical verb root across varying morphological forms (Gentner 2006: 549). The following quote should serve as a transition to the next section:

In semantic terms the grammatical morphemes [...] add number, tense, aspect, specificity or nonspecificity, containment or support. These modulations are inconceivable without the major meanings they modify and for this reason alone grammatical morphemes could not be acquired before content words and rules of combination and order. (Brown 1973: 389)

3. Acquisition of past tense structures

The past tense modification of English verbs signals two different meanings: first, what Brown calls “earlierness”, and second, to posit a substitute for reality as in if I were you (1973: 322). The following will only consider children’s acquisition of the semantic concept of earlierness, i.e., “the occurrence of an action or state at a time anterior to the time of utterance” (Brown 1973: 322). Brown investigated the speech of three preschool children—Adam, Eve, and Sarah—and found that earlierness is semantically understood from an early age (1973: 322). For example, Eve and Sarah, when they were two and a half years old, produced utterances such as What did you doed?, Did you bought this?, and I didn’t did nothing (Brown 1973: 333). Even though they were mistaken in signaling earlierness twice, none of these errors suggest a failure to understand the underlying semantic concept of the past tense form (Brown 1973: 333).

3.1 Tense and aspect

Verbs can be inflected for tense and aspect and are thus the major means of conveying temporal information” (McShane, Whittaker & Dockrell 1986: 275). Logically, in order for children to modify verbs according to the rules of word formation to add tense and aspect, they have to understand the notions of time concepts beforehand. Reichenbach studied children’s knowledge of tense and aspect in the 1940s and distinguished between speech time (ST) which refers to the temporal point at which the utterance takes place; event time (ET) which refers to the point at which the event described in the utterance takes place; and reference time (RT) by which pastness and futurity in relation to ST and ET are measured (Goodluck 1991: 129). In 1980, Smith proposes that in the early stages of speech development, children’s point of view is limited to the present, i.e., RT is always equivalent to ST (Goodluck 1991: 129). According to Smith, this stage will be over at the age of four at the latest (Goodluck 1991: 129). She provides evidence from an experiment in which English-speaking children aged four and older almost invariably used the simple past, e.g., He walked, and the past progressive, e.g., He was walking, to describe events acted out with toys (Goodluck 1991: 130). Thus, English-speaking children aged four years and older are capable of using past tense inflections in a manner that is similar to adult grammar, and, above all, their use of past tense inflections is an indicator that they are able to distinguish between the present and the past (Goodluck 1991: 130). On the cognitive level, Behrens argues, children must have some linguistically-independent temporal representations and concepts and, on the language level, linguistic knowledge of inflectional morphemes in order to understand and produce sentences with different grammatical tenses (2001: 450-451). Tomasello states that the acquisition of verbs is the crucial turning point in children’s transition to mature grammatical competence (1992: 7).

3.2 Regular and irregular verbs

“The cognitive processes underlying inflectional morphology have been, and continue to be, hotly debated. Much research has focused on the English past tense, for which verbs fall into two groups” (Marshall & van der Lely 2012: 122). Approximately 180 English verbs take an irregular past tense form, whereas the vast majority of English verbs are regular, i.e., they take the / ed / suffix (Plunkett 1995: 54). This certainly constitutes a challenge for English-speaking children because they have to learn which verbs are irregular and how these are conjugated through memorisation. The widely accepted model of the dual route seeks to explain the development of children’s acquisition of the English past tense:

First, a memory storage device contains the past tense of highly frequent and irregular forms in the language. Second, a rule-based system appends the appropriate allomorph of / ed / to the stem of the verb to form the past tense. Early correct usage of past tense forms is explained by the operation of the first memory storage device. The onset of overgeneralization errors is explained by the interference of the two mechanisms. Specifically, the memory storage device fails to block the application of the regular rule to an irregular stem. Finally, mature competence is explained by the two mechanisms discovering the correct division of verbs into regulars and irregulars. (Plunkett 1995: 38-39)

According to this model, early acquisition of the past tense is characterized by an error-free performance with children simply remembering and imitating the input of parental speech (often irregular and high-frequency verbs). After children have figured out the rule for the regular past tense, i.e., the addition of the / ed / suffix, they start to produce overgeneralized forms (such as goed, ated). Children have acquired the English past tense when the symbolic route for exceptions, i.e., for irregular verbs, is able to block the application of the regular / ed / suffix to an irregular stem. As shown by these examples, it is important to note that overgeneralization errors can not only occur with a wrong suffixation of the base form, e.g., goed, but also with a redundant suffixation of the correct past tense form, e.g., ated (Marshall & van der Lely 2012: 131). However, Hammer claims that the latter solely occurs in approximately 1-2% of children’s early past tense forms and that the first type of overgeneralization errors—the application of the regular suffix / ed / to the verb stem (such as eateated)—occurs far more often (2010: 125). Logically, since the memory trace is still weak within the preschool years, the more often children hear specific irregular past tense forms, the less likely it is that they will overgeneralize them (Hammer 2010: 133). This claim steps into the context of frequency. Interestingly, there seems to be no research data that account for the acquisition of the regular past before the irregular past. In fact, the majority of high-frequency verbs are irregular (Rumelhart & McClelland 1993: 538). Brown observed in his longitudinal study of the three preschool children Adam, Eve, and Sarah that their first past tense forms were came, broke, fell, sat, and went (1973: 260). Furthermore, he observes that irregular past tense forms were more frequent in every stage of language development than regular past tense forms—similar to adult speech (Brown 1973: 260). Additionally, Tomasello found that his daughter’s first past tense forms were all irregular: stuck, gone, broke, gave, did, drew, ate, found, fell, and had (1992: 166). Like most English-speaking children, Tomasello’s daughter uttered her first (irregular) past tense form rather early—with one and a half years (1992: 161). She learned over 150 verbs before her second birthday, and, subsequently, she started to produce overgeneralized past tense forms (Tomasello 1992: 219-221). The dual route model claims that there is an initial period in which English-speaking children produce the correct past tense form. Plunkett provides evidence for this since she argues that overgeneralization errors have not yet been observed until the vocabulary size reaches approximately 120 verbs (1995: 48). The advance of irregular past tense forms appears to be semantic. Events which are likely to happen just before an utterance carry the first past tenses (Potts, Carlson, Cocking & Copple 1979: 93). This is also evident in the findings mentioned above since Brown and Tomasello observed that broke, fell, stuck, came, and gave etc. were the first past tense forms of the children which were being studied. Therefore, it can be assumed that children hear these past tense forms only in those situations where the past tense refers to an “immediate past event which resulted in some present different end-state (e.g., the lamp broke)” (Kuczaj 1977: 595).

Kuczaj argues that children use irregular past tense forms such as broke as “syntactically unanalyzed but semantically appropriate forms” (1977: 598) because they have learned these past tense forms in those specific contexts rather than having analyzed that broke is the past tense form of break. Edwards and Goodwin propose that in the early stages, children are only capable of talking about physical events which are taking place around them (1986: 257). The fact that young children have probably not yet developed a representational system of the past which goes beyond the immediate context provides a reason why English-speaking children produce utterances with irregular change-of-state verbs first (Behrens 2001: 456). Evidence for these hypotheses can be found in Tomasello’s study. He observed that his daughter’s first verbs were change-of-state verbs which involved an object undergoing a transformation, e.g., fell or stuck, and activity verbs which in the broadest sense involve a temporal transformation, e.g., slept or swept (Tomasello 1992: 265). Overall, there seems to be a U-shaped development when it comes to children’s acquisition of past tense inflection (Hammer 2010: 126). This U-shaped development is a sequence of three stages (Goodluck 1991: 51). In the first stage, children learning English as their native tongue use the past tense forms of almost exclusively high-frequency irregular verbs; in this stage, there is no productive knowledge of the regular past tense rule (Rumelhart & McClelland 1993: 538). In the second stage, children know more regular past tense forms than irregular past tense forms and there seems to be implicit knowledge of a linguistic rule since overgeneralization errors start to occur (Rumelhart & McClelland 1993: 538). At this stage, children not only tend to incorrectly apply the regular past tense ending / ed / to irregular roots as in doed, or to apply this suffix to the irregular past tense form as in ated, they are also capable of generating a past tense form for made-up words (Rumelhart & McClelland 1993: 538). Rumelhart and McClelland mention a study of Berko in which children who were convinced that rick is used to describe a particular action produced ricked when they had to use the word in the past tense (1993: 538). In the third stage, overgeneralization errors and correct irregular past tense forms temporarily coexist until children systematically produce correct regular and irregular past tense forms (Rumelhardt & McClelland 1993: 538). This coexistence is relatively consistent from the age of three to at least into the sixth year (Hammer 2010: 127), but it has also been observed that overgeneralization errors persist into the tenth and eleventh years of life (Kuczaj 1977: 600). Several linguists have observed this kind of developmental pattern. For example, Eve, from Brown’s study, used came 11 times between 20 to 22 months and then started to use comed between 25 to 27 months (Hammer 2010: 126). In the last two stages, children might also use the correct past tense form of an irregular verb, the irregular past tense form + ed, and the irregular base + ed within the same dialogue (Rumelhart & McClelland 1993: 539). Thus, Rumelhart and McClelland emphasize that although the transition from one stage to the other is indeed incremental, the performance within the stages is extremely individual (1993: 539). Furthermore, since irregulars need to be learned from input, irregular past tense forms which are rarely used are more likely to be overgeneralized (Hammer 2010: 127). Goodluck argues that overgeneralization errors are unequivocal evidence that children are capable of figuring out a rule from the verbal input around them (1991: 51). Kuczaj claims that age-related differences tend to determine the types of overgeneralization errors which children produce (1977: 595). In Kuczaj’s study, fourteen children were observed. The nine youngest children showed a greater tendency toward producing errors of the irregular base form + - ed type (e.g., goed) than of the irregular past tense form + - ed type (e.g., wented), whereas the five older children showed a greater tendency to make more errors of the wented type than the goed type (Kuczaj 1977: 593-594). These findings match Hammer’s claim that a wrong suffixation of the irregular base form is the most common error in early acquisition (2010: 125). It is generally assumed that children learn inflected forms on an item-by-item basis, that is, as separate lexical entries (McShane, Whittaker & Dockrell 1986: 294). Thus, words such as caught, catching and catches are “unanalyzed wholes for the child and not the product of rules that combine the verb stem with an inflection” (McShane, Whittaker & Dockrell 1986: 294). This is why children typically use the correct word form in the early acquisition period until they discover regularities, e.g., adding / ed / to the verb stem when referring to the past (McShane, Whittaker & Dockrell 1986: 294). According to McShane, Whittaker, and Dockrell, this is done in the following manner:


Excerpt out of 13 pages


First Language Acquisition. How Englisch-speaking children acquire past tense structures
Christian-Albrechts-University of Kiel  (Philosophische Fakultät - Englisches Seminar)
From the Phoneme to the Word: Semantics
Catalog Number
ISBN (eBook)
ISBN (Book)
Semantics, Linguistics, First Language Acquisition, English, developmental linguistics, acquisition of past tense structures, past tense structures, tense and aspect, language acquisition, learning verbs, native tongue, Sprachwissenschaft, grammatical morphemes, English past tense, suffix, frequency, overgeneralization, regular and irregular verbs, grammatical development
Quote paper
Bachelor of Arts Jella Delzer (Author), 2020, First Language Acquisition. How Englisch-speaking children acquire past tense structures, Munich, GRIN Verlag,


  • No comments yet.
Look inside the ebook
Title: First Language Acquisition. How Englisch-speaking children acquire past tense structures

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free