Deduction or construction. SCOTT ROBINSON discusses the pros and cons of generative and usage-based approaches to language learning

Understanding how children come to acquire language has been a great source of debate for many years.  From the nature or nurture debate to the much more detailed ideas on whether children are an active or passive part of language development, there is a lot to be said.  When it comes to language acquisition there are two main approaches, generativist and usage-based. According to the generativist theory, “all human children innately possess a universal grammar, abstract enough to structure any language of the world” (M. Tomasello, K. Abbot-Smith, 2002, p.207).  There are two processes in this model – “acquiring all the words, idioms, and quirky constructions of the particular language being learned; and linking the particular language being learned to the abstract universal grammar” (M. Tomasello, K. Abbot-Smith, 2002, p.207). As this is allegedly innate, grammar develops continually, which is different to that proposed by of the usage-based theorists. Tomasello, who is one of the key exponents of the usage-based theory, points out that in fact “children are not very productive with their early language, suggesting that they do not possess the abstract linguistic categories and schemas necessary to effortlessly generate infinite numbers of grammatical sentences” (M. Tomasello, K. Abbot-Smith, 2002, p.207). These are the differences in the theories which are debated most.

Tomasello assumed a constructivist approach to language acquisition. Ambridge and Lieven (2011, p 123) claim that according to constructivists, “language is an inventory of constructions of various sizes and various levels of abstraction, each of which serves some communicative or socio-pragmatic function”. Constructivism argues language is likely to be acquired through a development of the understanding of forms (syntactic categories such as nouns and verbs) through the function of meaning.  In more detail, the usage-based theory focuses “on the specific communicative events in which people learn and use language” (M. Tomasello, 2000, p.61). He argues that sufficient input is required for a child to acquire language and detracts away from the notion of innateness. This is highlighted in his response to Lidz et al, who argued that input for children does not generate sufficient information to support unaided learning and attempts to support the contribution of innateness. In his response to Lidz et al’s  (2003) study, he highlights that in a forced choice situation, eighteen month old children think the phrase ‘another one’ goes best with another object, near identical to the one they’ve just seen other than one that is a different colour (M. Tomasello, 2004, p.140). By highlighting this he is trying to prove this shows nothing of a child’s understanding of the nested structure of noun phrases or innate linguistic knowledge, and that only non-linguistic experiences are shown in the study (M. Tomasello, 2004, p.140).

The usage based approach can be summarised in two principles, the first being ‘meaning is use’ which “represents an approach to the functional or semantic dimension of linguistic communication” (M. Tomasello, 2008, p.69). This idea stems from philosophers of language who wanted to combat the idea that meanings are things in themselves and focus on how people use linguistic conventions to achieve social ends. The second is ‘structure emerges from use’ which “represents an approach to the structural or grammatical dimension of linguistic communication” (Tomasello,2008, p.69).  The theory places emphasis on pre-linguistic communication. It is important to start by exploring the communicative function in the usage based view. Human infants communicate in rather sophisticated ways prior to acquiring any linguistic conventions: e.g. nearly all infants communicate by pointing before they acquire productive language. For Tomasello this suggests “that all human pointing and other gestures may already embody forms of social cognition and communicative motivation that are unique to the species, and that are necessary as a first step on the way to linguistic conventions both phylogenetically and ontogenetically” (M. Tomasello,2008, p.70). This stands opposed to the generativist approach, which because they believe it is innate, “universal grammar does not develop ontogenetically but is the same throughout the life span…” (M. Tomasello, 2002, p.207).

SCOTT ROBINSON, English Language undergraduate, University of Chester, UK


Ambridge, B., & Lieven, E. V. M. (2011). Child language acquisition: contrasting theoretical approaches. Cambridge: Cambridge University Press.

Tomasello, M., Abbot-Smith, K., (2002) A tale of two theories: response to Fisher’, Cognition 83, pp.207 – 214.

Tomasello, M. (2004) ‘Syntax or semantics? Response to Lidz et al.’, Cognition 93, pp.139-140.

Tomasello, M. (2000) ‘First steps toward a usage-based theory of language acquisition’, Cognitive Linguistics 11-1/2, pp.61-62.

Tomasello, M. (2008) ‘The Usage-based theory of language acquisition’. In Bavin, E, (Ed.) The Cambridge handbook of child language (2008), pp.69-88.


Is grammar squished into our head from birth or built up slowly from usage? HANNAH SADLER gets cerebral

Whether infants learn language from absorbing the chit-chat of their caregivers or from an innate mechanism in those complicated, squishy things in our heads has been an ongoing and baffling debate for centuries. If you take this question to the internet or to the books, they will both come back at you with ample theories and theorists. But before I try to steer you to the empiricist side of the debate, I feel that it is only fair I give you an overview of what the other side is saying about language learning.

Ambridge and Lieven (2011, p. 1) explain that whilst there are many competing proposals, each will generally be aligned with one of two major theoretical approaches. These are: (a) the nativist, generativist, Universal Grammar (UG) approach and; (b) the empiricist, constructivist, usage-based approach. The most appropriate term to use will depend on the precise nature of the proposal.

The (a) approaches are associated with likes of Chomsky who argued that speakers must possess a system or set of rules that allows them to understand and use grammar correctly (1959, cited by; Ambridge and Lieven, 2011, p. 105). He famously illustrated this point with the following pair of sentences:

Colorless green ideas sleep furiously.
* Sleep green colorless furiously ideas.

Although you have probably never been exposed to either of these sentences before due to their nonsensical nature (unless you are already familiar with this example), your knowledge of English still allows you to determine that the first sentence is grammatical (it is a possible sentence in English despite how bizarre it may sound), whereas the second is not (Ambridge and Lieven, 2011, pp. 104-105). However, I couldn’t agree more with Saxton’s view that ‘[i]f your eyebrows shoot up at this idea, you are not alone.’ He claims that there are many who find this notion deeply implausible (2010, p. 187).

Those in favour of the (b) approaches are not fond of grammar being innate. Instead they believe that it is possible to learn grammar. Ambridge and Lieven (2011, p. 126) explain that learning construction grammar is a gradual process. They propose that the SUBJECT VERB OBJECT transitive constructions do not need to be learned all in one go. The child will learn phrases such as ‘I’m kicking it’, ‘I’m hitting it’ and ‘I’m eating it’  directly from the input and will then schematize to form an ‘I’m ACTIONing it’ schema. Theorists refer to this notion as a lexically specific or item-based construction (Ambridge and Lieven, 2011, p. 126).

This takes me on to my favourite guy in the area of language acquisition, Tomasello. Tomasello is a key thinker when it comes to the usage-based approach to language acquisition. Saxton explains that within the usage-based approach, three stages of early multi-word construction are recognized: word combinations, pivot schemas and item-based constructions (2010, pp. 213-215). Tomasello explains these in detail in his book Constructing a Language: A Usage-based Theory of Language Acquisition (2003). Unlike Chomsky, Tomasello has put his ideas in writing in a fashion that has a fighting chance of keeping you awake.

So what is Tomasello saying about grammar? In terms of item-based constructions he explains the what and the why. What are infants doing? They are producing transitive utterances around their second birthdays which are verb specific. He claims that there is abundant evidence to support this and even refers to his own daughter, explaining that all of her early multi-word utterances revolved around specific verbs (2003, p. 117). However, some verbs were used in quite simple constructions (‘cut’____) and some in more complex frames of different types (‘Draw ____ ‘, ‘Draw ____ on ____’ , Draw ____ for ____’ , ‘____ draw on ____’) (Tomasello, 2003, p. 117). Why do infants do this? Because of what they are exposed to linguistically. The usage-based approach assumes that infants will most easily acquire the words and constructions that they hear most frequently (Ambridge and Lieven, 2011, pp. 2-3).

Unfortunately we may never have a certain, flawless answer to how we acquire language. However, to put language learning “on a par with the elements of our common nature that causes us to grow arms and legs rather than wings” (Chomsky, 1988, p. 4 cited by; Saxton, 2010, p. 187) seems a less logical, small-minded and archaic view to have in terms of language acquisition.

HANNAH SADLER, English Language undergraduate, University of Chester, UK


Ambridge, B., & Lieven, E. V. M. (2011). Child language acquisition: contrasting theoretical approaches. Cambridge: Cambridge University Press.
Saxton, M. (2010). Child language: acquisition and development. London: SAGE.
Tomasello, M. (2003). Constructing a language: A usage-based theory of language acquisition. London: Harvard University Press.

Grown-up grammars and creative constructions. RACHEL BREEDON explores modern theories of child language acquisition

Whilst helping children work their way through the early stages of language, we may not recognise the effort that little ones put in to making those wonderful half-formed sentences. They make it look easy, but whilst they’re babbling away, most of us are unaware of the controversy surrounding this verbal miracle. How exactly do these tiny humans piece together and learn a language so fast and efficiently?

It is apparent that the ability to learn language is innate for humans specifically. Without delving too deeply into the realm of animal linguistics, one characteristic that separates human language from animal languages is the ability to make brand new, novel utterances. To do this though, we need some sort of grammar to be able to combine words and express features such as tense (Ambridge & Lieven, 2011, p. 103). This is where theories of language acquisition generally divide into one of two approaches; generative or constructivist, each with their own ideas of what the nature of this grammar is and where it comes from.

According to generative grammar theory, all children have an innate universal grammar. They believe the capacity to learn language is ‘hardwired’ into our brains in what Chomsky calls a language acquisition device (LAD), explaining our instinct for learning language. Generativists also hold the continuity assumption (Pinker, 1984), which suggests that universal grammar is the same throughout people’s life-span, assuming young children have the syntactic competence of adults.  Acquiring language according to the generativist view can be simplified into two processes; learning the words and constructions of the language, then linking the learned language to the abstract universal grammar (Tomasello, 2002, p. 207). Taking an example from Ambridge & Lieven (2011, p. 123), to form the sentence ‘John kicked Bill’ a speaker would retrieve the verb ‘kick’ from their lexicon or mental dictionary and see that it needs a kicker (subject) and a kickee (object). They’d then insert ‘John’ and ‘Bill’ in their subject and object positions and voilà! Sentence formed according to generativists.

However, this theory relying on selection struggles to account for the formation of certain sentence types. For example, “he sneezed the napkin off the table” would mean you would have to have a caused-motion meaning in your mental dictionary for the verb ‘sneezed’, which most of us have only encountered in intransitive sentences with no object, such as ‘he sneezed’ (Ambridge & Lieven, 2011, p.125). We don’t often encounter someone sneezing something to a place, so how would a child work out what that means?

This is where the constructivist approach emerges. A construction according to Goldberg (1995, p. 4) is “a form-meaning pair such that some aspect of the form or some aspect of the function is not strictly predictable from the construction’s component parts”. In simpler terms, the construction or word pattern itself adds meaning. For example, in an English transitive construction such as the kicking example from before, a form (NOUN1 VERB NOUN2) is associated with a function (A acts upon B, causing B to be affected in some way) (Ambridge & Lieven, 2011, p. 124). From the components ‘kicked’, ‘Bill’ and ‘John’ alone it isn’t possible to tell who was the kicker and kickee, but when inserted into the construction the meaning emerges.

Once a child notices that certain forms are associated with particular functions, they are motivated to learn that construction, especially when talking about their own actions. SUBJECT VERB OBJECT transitive constructions can be learned bit by bit, as the first step is to learn simple phrases such as ‘I’m hitting it’ and ‘I’m eating it’ directly from adult input (Ambridge & Lieven, 2011, p. 125). From these, children form an I’m ACTIONing it schema into which they can insert any action they learn the lexical item for. These are item-based constructions and are easily learned from the input of others, relating to simple functions that the child understands. Tomasello found that most of children’s early linguistic competence is item-based and develops in a piecemeal fashion. He provides evidence to contrast the generativist continuity assumption, suggesting there is virtually no evidence of an innate, system-wide adult-like grammar in children and proposing his own usage-based theory of language acquisition (Tomasello, 2000, p. 209).

However, even that study received criticism (Fischer, 2002), and so the dispute continues. It seems obvious that item-based schemas such as ‘I’m ACTIONing it’ could generate children’s early sentences. The challenging part is deciding whether they do, or whether we’re born with these abstract rules and just have to match our experiences up with them. Either way, innate or acquired, baby humans certainly have a knack for grammar.

RACHEL BREEDON, English Language undergraduate, University of Chester, UK


Ambridge, B., & Lieven, E. V. M. (2011). Child language acquisition: Contrasting theoretical approaches. Cambridge: Cambridge University Press.

Fischer, C. (2002). The role of abstract syntactic knowledge in language acquisition: a reply to Tomasello (2000). Cognition, 82, pp. 259-278.

Goldberg, A. E. (1995). Constructions: A construction grammar approach to argument structure. University of Chicago Press: Chicago.

Pinker, S. (1984). Language learnability and language development. Harvard University Press: Cambridge, MA.

Tomasello, M. (2000). Do young children have adult syntactic competence? Cognition, 74, pp. 209-253.


Blank slate or language acquisition device. MADDI SYMES explores the language acquisition battleground

From hair colour to height, gender to complexion – we all have characteristics which have been determined biologically. Whilst these characteristics are undeniably innate, there are many human traits which are not quite so straightforward; aggression, intelligence, our unique ability to communicate through language. This longstanding debate had existed since Francis Galton coined the phrase ‘nature versus nurture’ in 1869 (Cacioppo & Freberg, 2013, p. 90). Human beings have come so far in scientific discoveries, from black holes to cancer treatments, so why is it we still don’t know whether language is innate or learnt? Well, allow me to break it all down for you!

Rationalist René Descartes (1641; cited by Hunt, 2003, p. 32) argued that we have ‘innate ideas’ whilst empiricist John Locke (1689; cited by Sherman, 2013, p. 26) famously attacked Descartes claim, stating that the human mind ‘begins’ or enters the world in a blank state (tabula rasa) – knowledge being acquired through posteriori – through experience and observations. From an epistemological point of view, we have capacities to acquire language but clearly have no ability to develop it as we do this in early in life. Tabula rasa, although tenable in this sense, is simply not true. In the light of contemporary science it is evident that we begin life with certain characteristics that characterise ‘human nature’. Evolutionary, biological instincts and that are studied within the discipline of ‘evolutionary psychology’.

When exploring this debate with regards to language we find there are two major schools of thought – nativists and functionalists. Nativists believe that some aspects of linguistic knowledge are innate, meaning they are present at birth (Ambridge & Lieven, 2011, p. 1-3). Holding up a fight for nativism is generativist Noam Chomsky (1976). Generativists believe knowledge of grammar consists of formal ‘rules’ that operate on abstract linguistic categories. Chomsky (1976; cited by Kearns, 2010, p. 174) argues that these rules are innate and that we all have inbuilt, instinctive ‘universal principles’ and rules for grammar. He calls this our ‘Language Acquisition Device’ (LAD) which is activated when children are exposed to language (Chomsky, 1976; cited by Kearns, 2010, p. 174). If only it was that simple, Chomsky!

On the other side of the battle ring is Tomasello (2005; cited by Workman and Reader, 2014, p. 303), a leading figure in the functionalist camp. Functionalists argue that the ability to learn language is innate, but there is no innate knowledge of grammar, and grammatical categories are not a priori. Tomasello, a bit like Locke, believes we learn through exposure to adult speech.  A quote by Tomasello (2008) that I came across reading his chapter, ‘The Grammatical Dimension’ In Origins of Human Communication reads, “[a]lthough many aspects of human linguistic competence have indeed evolved biologically, specific grammatical principles and constructions have not” (p. 313) which I feel summarises the functionalist view well.

So far it seems to be a tie between the two contenders, so let’s step out of the battle ring and look at an experiment famously known as ‘the wug test’. I first came across the research a number of years ago when I was reading Davidson’s (2011) book Planet Worda great read by the way! Jean Berko Gleason (1958; cited by Davidson, 2011, p. 47) presented children with a picture of a ‘wug’, a nonsense word describing the creature in the picture to see whether the children made ‘wug’ into the plural ‘wugs’. The results suggested children’s ability to form such grammatical structures is varied and depends on the individual’s development (1958; cited by Davidson, 2011, p. 49). Nativists argue that this shows children do not simply imitate language as they can produce the correct grammatical forms for nonsense words they have not heard before (Barry, 2002, p. 184). However, there have been many criticisms of the ‘wug’ test, and could it be possible that children have actually acquired these grammatical rules and have learnt to generalise them?

Gleason has quite a middle ground opinion in believing that there are areas of the brain which are specialised for language and through hearing and experiencing language and by interacting with language users, coupled with the capacity for language, language is built in the brain (1958; cited by Davidson, 2011, p 49). Personally, I have to agree with Gleason’s (1958) view. I believe that the difficulty in concluding the on-going debate lies in the fact we cannot (yet) conduct experiments on new-born babies that would provide substantial evidence to prove which aspects of language are innate and which are acquired. Will it take breakthrough research to end this battle, or is it simply unanswerable?

MADDI SYMES, English Language undergraduate, University of Chester, UK


Ambridge, B., & Lieven, E. V. M. (2011). Child Language Acquisition: Contrasting Theoretical Approaches. Cambridge, United Kingdom: Cambridge University Press.

Barry, A. (2002). Linguistic Perspectives on Language and Education. Connecticut, United States of America: Greenwood Publishing Group.

Cacioppo, J., & Freberg, L. (2013). Discovering Psychology: The Science of Mind. Belmont, United States of America: Wadsworth.

Chomsky, N. (1976). On the nature of language. In S. R. Harnad, H. D. Steklis, & J. Lancaster (Eds.), Origins and Evolution of Language and Speech, 280, 46–57.

Davidson, J. P. (2011). Planet Word. London, United Kingdom: Michael Joseph.

Descartes, R. (1641). Meditations on first philosophies. In J. Cottingham, R.Stoothoff, & D. Murdoch (Eds.), The Principle Writings of Descartes. Cambridge, United Kingdom: Cambridge University Press.

Galton, F. (1869). Hereditary Genius. London, United Kingdom: Macmillan and   Co.

Gleason, J. (1958). The child’s learning of English morphology. Word, 14, 150–177.

Hunt, S. (2003). Controversy in Marketing Theory: For Reason, Realism, Truth, and Objectivity. New York, United States of America: M. E. Sharpe.

Kearns, K. (2010). Frameworks for Learning and Development. Australia: Pearson.

Locke, J. (1689). An Essay Concerning Human Understanding. London, United Kingdom: William Tegg.

Sherman, P. (2012). John Locke: Philosopher of the Enlightenment. California, United States of America: Teacher Created Materials.

Tomasello, M. (2008). Origins of Human Communication. London, United Kingdom: The MIT Press.

Workman, L., & Reader, W. (2014). Evolutionary Psychology (3rd ed.). Cambridge, United Kingdom: Cambridge University Press.


Will we ever solve the language acquisition puzzle? LAURA TALBOT investigates

Language sets us apart. Animals communicate but they do not have anything approaching the sophisticated grammar of human languages. How is it that we learn to speak and think in language so easily? By the time we are five years old we have an, ‘[e]stimated vocabulary of 5,000-10,000 words’ (Anglin, 1993, p.147). But how do we acquire them? Young children become adept in a new language very quickly and since the dawn of philosophy, thinkers have argued about whether or not they are born with innate structures to prepare them for the task. So are children born with blank slates in terms of language or do they have an outline knowledge of how language is structured?

Two key theorists within this debate are nativist Noam Chomsky, who believes that “[c]ertain aspects of our knowledge and understanding are innate, part of our biological endowment, genetically determined, on a par with the elements of our common nature that causes us to grow arms and legs rather than wings” (Chomsky, 1988, p.4). Suggesting that just as children are born ‘innately’ with arms and legs, they are also born with innate structures that allow them to understand and acquire language. The opposing theorist is, constructivist Michael Tomasello, who claims that, “children acquire language first and foremost by understanding how others use language” (Tomasello, 2009, p.86). This posits that children understand and acquire language through the nurturing of the people surrounding them.

Looking firstly at the nativist approach, the key question is whether or not we have ‘innate principles’. Nativists claim that the rules for sentence structure are too complex to be acquired by a learner who comes to the task with no knowledge of the way that language is structured, hence Chomsky’s theory of ‘innate principles’. He believes that “the child is born with some innate principles about language ‘wired in’ to the brain” (Chomsky 1981 as cited in Cattel, 2007, p.82). He posits that this basic innate knowledge of sentence structure and grammar is present in all human children- hence it is a ‘universal grammar’. He states that “[UG] may be thought of as some system of principles, common to the species and available to each individual prior to experience” (Chomsky, 1981, p.7). Could this be possible, when all languages are made up of different grammar rules and structures? Chomsky argues that “individuals in a speech community have developed essentially the same language” (Pinker, 1994, p.23). Nativist Steven Pinker uses the following example to advocate Universal Grammar, claiming “[w]e think children pick up their mother tongue by imitating their mothers, but when a child says Don’t giggle me! or We holded the baby rabbits it cannot be an act of imitation” (Pinker, 1994, p.21). This coincides with the idea that children create their own sentences using their own rules and innate ‘Universal Grammar’ to guide them.

In his book The ‘Language Instinct Debate’ (2005) Geoffrey Sampson directly disputes Chomsky and Pinker’s theory of ‘Language Universals’. He claims that “Nativists have often made authoritative-sounding claims about language universals when it is clear that the claims could not have survived a minimal attempt to check for counter-evidence” (Sampson, 2005, p.138). Sampson believes that nativists, like Chomsky, make claims about language universals and base it on one example, without using further examples or evidence.

So what about the major influence of the child’s surroundings? Surely their environment growing up affects their language?

This takes us to the constructivist approach of language acquisition, which argues that, “[y]oung children must learn during their individual ontogenies the set of linguistic conventions used by those around them” (Tomasello, 2003, p.1). We spend the majority of our first three years of life surrounded by adults, so it makes sense that we are going to pick up their ‘linguistic conventions’ and build upon them. Constructivists fundamentally believe that language develops alongside and in connection with other social and cognitive skills which develop at, “around 9-12months of age” (Tomasello, 2003, p.3). So children will acquire language as they acquire the ability to interact in communication, make gestures, point, share the intentions of others and pursue shared goals. Therefore, language acquisition is part of the development of cognitive abilities and the knowledge of language is arguably derived from experience of the outside world.

After thousands of years of debate, will we ever get a definite answer as to how we acquire language? Aitchison attempts to answer the question by stating that “everyone agrees that human beings bring some innate faculties to the task of learning, and everyone agrees that the mature human’s cognitive world depends in some respect on his experience’ (Aitchison, 2011, p.28). Will theorists like Tomasello and Chomsky eventually join forces and reach a conclusion? The debate continues….

LAURA TALBOT, English Language undergraduate, University of Chester, UK


Aitchison, J. (2011). The Articulate Mammal: An Introduction to Psycholinguistics. London: Routledge.

Anglin, J.M. (1993). “Vocabulary Development: A Morphological Analysis.” Monographs of the Society for Research in Child Development 58(10 (238)): 1-166.

Cattell, R. (2007). Children’s Language: Consensus and Controversy. London: Continuum.

Chomsky, N. ‘Review of Verbal Behaviour by B.F. Skinner’. Language 35 (1):26-58.

Pinker, S. (1994). The Language Instinct. London: Penguin.

Sampson, G. (2005). The ‘Language Instinct’ Debate. London: Continuum.

Tomasello, M. (2003). Constructing A Language: A Usage-Based Theory of Language Acquisition. Cambridge, MA: Harvard University Press.