Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

Contents lists available at ScienceDirect

Neuroscience and Biobehavioral Reviews

journal homepage: www.elsevier.com/locate/neubiorev

Review article

Language acquisition from a biolinguistic perspective

Stephen Crain a,b,∗ , Loes Koring a,c , Rosalind Thornton a,b
a
ARC Centre of Excellence in Cognition and its Disorders, Australia
b
Department of Linguistics, Macquarie University, Sydney, Australia
c
Department of Cognitive Science, Macquarie University, Sydney, Australia

a r t i c l e i n f o a b s t r a c t

Article history: This paper describes the biolinguistic approach to language acquisition. We contrast the biolinguistic
Received 2 March 2016 approach with a usage-based approach. We argue that the biolinguistic approach is superior because it
Received in revised form 28 June 2016 provides more accurate and more extensive generalizations about the properties of human languages,
Accepted 8 September 2016
as well as a better account of how children acquire human languages. To distinguish between these
Available online 12 September 2016
accounts, we focus on how child and adult language differ both in sentence production and in sentence
understanding. We argue that the observed differences resist explanation using the cognitive mecha-
Keywords:
nisms that are invoked by the usage-based approach. In contrast, the biolinguistic approach explains
Biolinguistics
Language acquisition
the qualitative parametric differences between child and adult language. Explaining how child and adult
Unification language differ and demonstrating that children perceive unity despite apparent diversity are two of the
Universal Grammar hallmarks of the biolinguistic approach to language acquisition.
Structure-dependence © 2016 Elsevier Ltd. All rights reserved.
Continuity Assumption
Usage-based approach

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
1.1. The usage-based approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
1.2. Elaborating the biolinguistic approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
2. Structure dependence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
2.1. An account of structure dependence based on functional units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
2.2. Displacement in human languages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
2.3. Children as distribution analyzers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
2.4. Structure-dependent interpretations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
3. Anaphoric relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
3.1. Blocking coreference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
3.2. Thornton et al. (2015) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
3.2. Language delay: illicit coreferences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
3.3. Guasti and Chierchia (2000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
4. Children’s long distance wh-questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
4.1. Wh-copying by English-speaking children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
4.2. Children’s long distance wh-questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
4.3. A usage-based account of children’s non-adult utterances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5. Negation in child language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.1. A usage-based account of the acquisition of negation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.2. A biolinguistic account of the acquisition of negation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
6. Scope relations in human languages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
6.1. A substitute for negative evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
6.2. The isomorphism hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140

∗ Corresponding author at: Department of Linguistics, Australian Hearing Hub, 16 University Avenue, Macquarie University, NSW 2109, Australia.
E-mail address: stephen.crain@mq.edu.au (S. Crain).

http://dx.doi.org/10.1016/j.neubiorev.2016.09.004
0149-7634/© 2016 Elsevier Ltd. All rights reserved.
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 121

6.3. A challenge to the isomorphism hypothesis: reconstruction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

6.4. A second challenge: raising . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
6.5. Domain specificity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
6.6. How languages differ in scope assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
6.7. The Disjunction Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
6.8. A linguistic universal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
7. Complete nature as different aspects of one set of phenomena . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
7.1. Disjunction as an existential expression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
7.2. Free choice inferences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
8. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148

1. Introduction that all typically-developing children internalize a rich and com-

plex linguistic system in just a few years. Acquisition of language
There are many ways we could start this chapter, but a good is rapid and effortless for children, according to the biolinguis-
place to start is with the Modularity Hypothesis. The Modularity tic approach, because the acquisition of language builds upon a
Hypothesis supposes that the human mind/brain is comprised of foundation that is pre-determined by the biological endowment
“separate systems [i.e., the language faculty, visual system, facial of the species. The human biological endowment for language is
recognition module, etc.] with their own properties” (Chomsky, called Universal Grammar. Universal Grammar is the initial state
1988). Proposals about the nature of modularity differ in at least of the language acquisition device (LAD). Universal Grammar con-
two important respects. First, modular systems can be restricted tains core principles that are common to all human languages but,
to perceptual processes, or they can be taken to also encompass in addition, it contains information about ways in which human
higher-level cognitive abilities, such as language and reasoning. A languages differ. Information about language variation is encoded
second difference concerns whether modular systems are innate, in parameters. Universal Grammar, then, is a system of principles
or become ‘automatized’ through experience. Although modular- and parameters. Although the principles of Universal Grammar are
ity does not entail the innateness of cognitive systems (see e.g., inviolable, children use triggering experience to set the parame-
Karmiloff-Smith, 1992), most proponents of modularity advocate ters of Universal Grammar in order for children to adopt the same
some version of the innateness hypothesis. All advocates of modu- parameter values as adult speakers of the local language. Before
larity share one assumption, that of domain-specificity. A module certain parameters are set to the values adopted by the local lan-
operates on objects in a specific domain. In the Modularity of Mind guage, however, the language spoken by children can differ from
(1983, p. 51) Fodor asserts that “. . .the perceptual system for a lan- the language spoken by adults in the same linguistic community.
guage comes to be viewed as containing quite an elaborate theory of Such differences are nevertheless highly circumscribed. Essentially,
the objects in its domain; perhaps a theory couched in the form of a child language can differ from the language spoken by adults only
grammar of the language.” The focus of the biolinguistic program is in ways in which adult languages can differ from each other. This is
on language as a modular perceptual system. More specifically, the called the Continuity Assumption (Crain, 1991; Pinker, 1994; Crain
biolinguistic program is concerned with how sentences and their and Pietroski, 2001). The Continuity Assumption is one of the main
associated meanings are acquired by children, how they are used topics of this chapter.
by both children and adults, how the system that pairs sentences
and their meanings evolved, and how this system is represented in
the mind/brain. 1.1. The usage-based approach
There is now considerable empirical evidence that language has
the status of a module. The evidence takes several forms, includ- The usage-based approach to language acquisition stands in
ing the fact that (a) any human language can be rapidly acquired stark contrast to the biolinguistic approach. There is nothing
by any typically-developing child in the absence of decisive envi- approaching the Continuity Assumption according to the usage-
ronmental data, (b) language is unique to humans, (c) language based approach. Rather, this approach supposes that children
shows neurological localization from birth, and (d) language can be accrue linguistic knowledge in response to environmental input,
selectively impaired in special populations including some forms of using domain-general learning mechanisms, such as analogy and
brain damage and some genetic childhood disorders, and (e) lan- distributional analysis (Lieven and Tomasello, 2008; Saxton, 2010).
guage acquisition is governed both by a critical period and by a Initially, linguistic knowledge is accrued in a piecemeal fashion. The
maturational timetable. The present study describes the biolinguis- products of language learning, including the generalizations that
tic approach to language acquisition. Chomsky (2007, p. 2) states older children form, consist of ‘shallow’ records of their linguistic
the task as follows: experience (see e.g., Pullum and Scholtz, 2002). The linguistic sys-
tem that children internalize consists of constructions (templates,
“In biolinguistic terms, that means discovering the operations
schemas, constructs) (see Goldberg, 2003, 2006). For this reason,
that map presented data to the I-language attained. Abstractly
many advocates of the usage-based approach call themselves con-
formulated, it is the problem of constructing a ‘language acquisi-
structivists.
tion device’ (LAD), the problem of ‘explanatory adequacy’. With
A basic tenet of the usage-based approach is the claim that more
sufficient progress in approaching explanatory adequacy, a fur-
frequent constructions are mastered earlier in the course of lan-
ther and deeper task comes to the fore: to transcend explanatory
guage development than less frequent ones (Ambridge and Lieven,
adequacy, asking not just what the mapping principles are, but
2011; Lieven and Tomasello, 2008). Given that constructions are
why language growth is determined by these principles rather
initially acquired piecemeal, children are expected to take a con-
than innumerable others that can easily be imagined.”
siderable time to internalize a system that pairs utterances and
One of the most basic observations underpinning the biolinguis- meanings in the same way as adult speakers. Moreover, when
tic approach to language acquisition is the naturalistic observation children start to form generalizations that extend beyond their
122 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

experience, at around 4- to 5-years of age, the generalizations they from the surface. On the biolinguistic approach, it is likely therefore
form are just instances of a completely general problem of induc- to turn out that what are considered to be different construc-
tion. Learning to project beyond one’s linguistic experience is seen tions on the usage-based approach draw upon the same principles
to be just one variant of the problem that arises for learning all sorts of Universal Grammar. Because disparate-looking phenomena are
of things. derived from the same principles, children acquire these phenom-
As noted earlier, one of the main issues we will be concerned ena in concert, rather than piecemeal. This explains why language
with is the nature of the differences between child and adult lan- acquisition is so rapid and effortless for children, who master even
guage. According to the usage-based account, before children have seemingly complex structures by the age of 3.
identified the form-function mappings of the local language, they The biolinguistic approach offers an explicit account of the (lim-
are expected to produce less articulated versions of the construc- ited) ways in which child and adult languages can vary. This feature
tions that are produced by adults, missing certain of the linguistic of language acquisition is explained, in part, by the parameters of
ingredients that are present in adult speech. As children take on Universal Grammar. Just as parameters determine, at least in part,
board more and more constructions, child language is expected how adult languages differ from each other, parameters are also
to more closely match that of adults. Therefore the usage-based invoked to explain children’s non-adult linguistic behaviour. This
approach can be characterized as an “input matching” process. As is stated as the Continuity Assumption, which maintains that chil-
Lieven and Tomasello (2008, p. 171) remark: dren’s non-adult linguistic behaviour follows the natural seams of
human languages. To cite an example we will return to later, some
“The difference between young children’s inventories and those
languages permit an overt (phonetically realized) copy of a wh-
of adults is one of degree: many more, initially all, of chil-
phrase in the middle of wh-questions. In other languages, inserting
dren’s constructions are either lexically-specific or contain
an extra wh-word renders such questions unacceptable, as in What
relatively low-scope slots. As well as being less schematic than
do you think what Bill wants to do? In keeping with the Continuity
many adult constructions they are also simpler with fewer
Assumption, some English-speaking children initially produce wh-
parts. And, finally, children’s constructions exist in a less dense
questions with an extra copy of the wh-word, so children produce
network—they are more “island-like”.”
questions that are acceptable in some languages, but not in the local
The usage-based approach adopts the view that meaning is use, language (Thornton, 1990). The finding that children add structure
where “the primary psycholinguistic unit of child language acquisi- is consistent with the Continuity Assumption, but it is not consis-
tion is the utterance, which has as its foundation the expression and tent with the usage-based approach, which contends that children’s
understanding of communicative intentions” (Tomasello, 2000b; p. non-adult utterances should be “simpler with fewer parts.”
61) What children acquire, then, is a mapping of forms with func- Of course, the omission of linguistic material is also consistent
tions. The usage-based account purports that, in tandem, form and with the Continuity Assumption. For example, it has been well
function also explain how children build up relations among con- documented that English-speaking children sometimes omit entire
structions. As children progress towards the final stages of language noun phrases that are required to be phonetically realized by adult
development, they form abstract semantic relations among con- speakers. A parade case of this is a stage at which English-speaking
structions. The final stage of language development is outlined as children omit Subject noun phrases (e.g., Hyams, 1986). Although
follows by Lieven and Tomasello (2008, p. 171): many languages optionally omit Subjects (e.g., Spanish, Italian,
Mandarin Chinese), the kinds of omissions children make are unac-
“Finally, the child has to abstract the relations between con-
ceptable for adult speakers of English. Again, child English differs
structions. Evidence that this has occurred is that the child is
from adult English in ways in which adult languages differ from
able to transform an utterance in one construction into another
each other. We will also report the findings of several experimen-
construction, for instance a declarative into a wh-question or an
tal studies showing that children assign non-adult interpretations
active into a passive. This could be done by forming a semantic
to certain sentences. Again, cross-linguistic research reveals that
representation of what the speaker wishes to say, thereby allow-
children’s non-adult interpretations are licensed in possible human
ing the production of the other construction. Whether and when
languages, but not in the language spoken by adult members of the
the learner actually maps the form—function mappings of one
community in which the child is being raised. This position was first
construction to those of the other is an empirically open ques-
formulated in Chomksy (1965); see also Pinker (1984) and see Yang
tion at the moment. It depends on the metalinguistic expertise
(2002) for a formal implementation of this approach to language
and/or educational level of different speakers.”
learnability.
As this quote indicates, the usage-based approach is open to In all of these cases, children’s non-adult linguistic behaviour is
findings showing that different people develop different proficien- a consequence of the fact that they initially adopt different val-
cies in language. ues of parameters than those adopted by adult speakers of the
local language. Children’s initial non-adult assignments of param-
eter values do not impede their language acquisition, however. In
1.2. Elaborating the biolinguistic approach each case, children’s initial setting of parameters conforms to a
learnability mechanism known as the Subset Principle (Berwick,
In contrast to the usage-based approach, the biolinguis- 1985). This mechanism ensures that children have readily avail-
tic approach contends that language acquisition is rapid and able ‘positive’ evidence informing them that they need to ‘reset’
effortless, because language learners come equipped with the prin- the relevant parameters to the values adopted by the local lan-
ciples and parameters of Universal Grammar. Children do not guage. This evidence takes the form of ‘detectable errors,’ i.e., forms
acquire constructions one-by-one. Rather children amalgamate or meanings that the child’s grammar cannot generate using the
even disparate-looking linguistic phenomena, where these draw child’s current grammar. The fact that child and adult language dif-
upon the same principles of Universal Grammar. Principles of Uni- fers in non-trivial respects is not expected to hinder children from
versal Grammar apply within individual languages, tying clusters rapidly converging on a grammar that is equivalent to that of adult
of phenomena together. And these same principles apply across speakers.
languages, tying together similar phenomena in even historically It has been shown by advocates of the usage-based approach
unrelated languages. Uniting phenomena within and across lan- that children’s productions represent only a small proportion of all
guages requires principles that operate at a considerable distance of the possible syntactic combinations of certain word sequences.
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 123

According to the usage-based approach, the finding that children’s

sentences are “island-like” reflects the statistical distribution of
sequences of words in the input children encounter (cf. Tomasello,
2003). Even taking the findings at face value, the conclusion reached
by advocates of the usage-based approach is unwarranted. The fact
that children’s productions lack broader statistical coverage, con-
sidering all of the syntactic combinations that are logically possible
in adult language, does not entail that children’s productions are
not rule-governed (Valian et al., 2009; Yang, 2013). In this regard, Fig. 1. Structure-dependent Yes/No Questions.
it is worth pointing out that Valian et al. (2009) empirically demon-
strated that child and adult language do not differ significantly
in combinatory diversity. And Yang (2013) has demonstrated that 2008; pp. 245–248) and claim that “children acquire questions
(due to Zipf’s law) the observed diversity in children’s productions as an independent construction.” According to the biolinguis-
is more accurately modeled by a rule-based grammar than by mod- tic approach, all linguistic behaviour, including the formation of
els that rely on memorization and recall of word combinations. Yes/No questions, adheres to structure-dependence. The biolin-
Finally, according to the biolinguistic approach, all typically- guistic approach contends that all derivations by children and
developing children converge on a linguistic system that is adults, across languages, are structure-dependent – the mind
equivalent to that of adult speakers of the local language. Because imposes structure onto experience, and not the other way around.
the human faculty for language is viewed as a domain specific Note, however, that a computationally simpler, structure-
perceptual system (i.e., a module), this approach contends that independent operation could also derive (2) from (1). The
all children come to the task of language acquisition armed with structure-independent operation simply treats sentences like
the principles and parameters of Universal Grammar. The linguis- beads-on-a-string. The operation proceeds from left-to-right, one
tic abilities of language learners are not expected to depend on a word at a time, until it encounters a member of a list of words {is,
person’s level of education, for example. The principles and param- can, will,. . .}. When it finds one of these words, the word is reposi-
eters of Universal Grammar explain children’s convergence on the tioned at the beginning of the sentence. The structure-dependent
grammar of the local language before the age at which they begin rule (SD) and the structure independent operation (SI) are summa-
to receive formal education. By 3-years-old, children are effectively rized in (3).
adults in their abilities to produce and understand sentences they
have never encountered before, to judge the truth or falsity of these
sentences, and to discern entailment relations between them (see,
e.g., Crain and Thornton, 1998, 2015).
The sections that follow report the findings of experimental
studies of child language that reveal young children’s knowledge
of a rich and complex grammatical system. We chose these stud- Both the structure-independent operation and the structure-
ies because they focus on topics that have been investigated both dependent rule are compatible with the majority of the input
by researchers who adopt the biolinguistic perspective and by available to young children (Chomsky, 1980). The biolinguis-
researchers who adopt the usage-based perspective. The findings tic approach nevertheless proposes that children never adopt
of these studies therefore allow us to compare the empirical and a structure-independent operation, according to which children
explanatory adequacy of both approaches to language acquisition. treat sentences as strings of words. On this approach, all linguis-
These studies were selected for two other reasons. First, they are tic rules that child language learners formulate and execute are
experimental investigations of linguistic phenomena that children ones that analyse sentences into hierarchical structures. Children
master before they reach school age, almost without exception.1 are precluded by human biology from treating sentences as strings.
Second, they are investigations of linguistic structures that children Therefore the biolinguistic approach predicts that children will not
acquire in stages, including stages at which children produce non- commit the kinds of errors that would result from the application of
adult sentences or assign non-adult interpretations to sentences. structure-independent operations. That is, children will never pro-
duce sentences using structure-independent operations, despite
the simplicity of such operations and their ability to replicate the
2. Structure dependence
input children encounter.
The empirical coverage of the structure-dependent rule and
Consider examples (1) and (2). Example (2) is a Yes/No question,
the structure-independent operation is roughly the same when
and (1) is its declarative counterpart. On the biolinguistic approach,
the sentences under consideration are simple, as in examples (1)
these two sentences are related. The Yes/No question (2) is trans-
and (2). However, the superiority of the structure-dependent rule
formed from the declarative sentence (1) by a rule. Essentially, the
becomes visible when the sentences that are under consideration
rule moves the copula verb is from its sentence-internal position in
are more complex. One kind of sentence that reveals the superi-
(1) to the sentence-initial position in the Yes/No question (2).
ority of the structure-dependent rule is illustrated in (4). Sentence
(4) contains both the auxiliary verb ‘is’, which appears inside the
relative clause ‘the dog that is sleeping’, and another instance of the
same word (the copula verb is) in the main clause. The Yes/No ques-
tion that results from the application of the structure-dependent
rule to the declarative sentence (4) is given in (5), and illustrated
Advocates of the usage-based approach “do not accept the claim in the hierarchical structure shown in Fig. 1. The Yes/No ques-
that questions are formed by a movement rule” (Ambridge et al., tion that results from the application of the structure-independent
operation is given in (6). Movement of the first verbal expression,
from inside the relative clause, results in a clearly ill-formed Yes/No
1
See Section 3.2 for a discussion of the finding that children appear to be delayed question. Assuming that children never formulate structure inde-
in the acquisition of one linguistic phenomenon. pendent operations, they are never expected to produce Yes/No
124 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

questions like (6), even when they first start to produce Yes/No
questions that contain relative clauses.

This empirical prediction was first tested in an experimen-

tal study reported in Crain and Nakayama (1987). The Crain and
Nakayama study elicited Yes/No questions from 3- to 5-year-old Fig. 2. Forming wh-questions.
English-speaking children. The child’s task was to pose questions
to a character from Star Wars, Jabba the Hutt. The experimenter independent errors to extend beyond this construction. If the
explained that this would be a good test to see if Jabba the account is extended, the generalization would be as follows: Chil-
Hutt could speak English. The declarative sentences, with relative dren construct strings of words that form referential units, and do
clauses, were presented to the child in carrier phrases of the form not extract words from sentences if the result would break up a
“Ask Jabba if . . .”. To elicit the Yes/No question corresponding to (4), “functional unit,” i.e, strings of words that can be “used to make
for instance, the experimenter requested the child to “Ask Jabba if an act of referring.” This extended generalization is clearly false.
the dog that is sleeping is on the blue bench”. Children produced To see why, consider the example in (7). In the example, accord-
many different kinds of Yes/No questions. About 60% of children’s ing to the biolinguistic approach, an expression has been displaced.
Yes/No questions were adult-like questions. The remainder differed A movement rule has displaced the wh-word what. Although this
in certain ways from those of adults, so it was clear that children word appears in sentence-initial position, it was initially part of a
were at the early stages in forming the Yes/No questions associ- referential NP that itself contains a wh-word, a book about what. Fol-
ated with sentences that contained relative clauses. Nevertheless, lowing the displacement of what, the string of words that remains,
the child participants in the Crain & Nakayama study never pro- a book about, cannot be used to perform an act of reference (Fig. 2).
duced any Yes/No questions that could be characterized as ones that
had been derived from a structure-independent operation. Chil-
dren never posed Yes/No questions like (6). Crain and Nakayama
(1987) took this as evidence that children never hypothesize that a
structure-independent operation is the source of Yes/No questions.

The property of displacement is common in human languages.

2.1. An account of structure dependence based on functional units
In English wh-questions, the displacement of part of a wh-phrase
often leaves a preposition behind, at the end of the question, e.g.,
Advocates of the usage-based approach have disputed the con-
Which book did you find the answer in []? or Where does bacon come
clusion reached by Crain and Nakayama (1987). We will discuss
from []? These wh-questions end with the ‘stranded’ prepositions in
their objections momentarily, and point out why these objections
and from. This phenomenon is called ‘preposition stranding.’ Prepo-
are unwarranted. First, we wish to consider an alternative to the
sition stranding is highly preferred in colloquial English. In some
structure-dependent (SD) rule in (3). This alternative was intro-
languages, however, preposition stranding is not tolerated; prepo-
duced by Tomasello (2008, p. 85), who argues that children only
sitions must be moved along with the remainder of the expression.
appear to be using a structure-dependent rule. What children are
This is referred to as pied piping – a reference to the Pied Piper
actually doing is simply maintaining the integrity of the string
of Hamelin, who used a flute to lure away rats, and later children,
of words, the dog that is sleeping. According to Tomasello (2008,
away from the town. The suggestion by Tomassello that functional
p. 85), children refrain from extracting the auxiliary verb is from
units should remain intact implies that children should consistently
this string; otherwise, the remaining string, the dog that sleeping,
prefer pied piping to preposition stranding, just as some languages
would not serve its referential function:
do. If this were the case, then English-speaking children would pro-
“If children understand NPs with relative clauses − if they duce questions like In which book did you find the answers? and From
understand that the whole phrase is used to make one act of ref- where does bacon come? Clearly, these are not the kinds of questions
erence − then there would never be any temptation to extract children actually produce.
an auxiliary from it; they would simply understand that that Although preposition stranding is preferred to pied piping in
unit stays together as one functional unit.” most wh-questions, English avoids a similar decoupling in wh-
questions with the wh-word whose, e.g., Whose book did he buy?
In sentence (5), on the other hand, the string of words, the dog
and Whose book do you think he bought? In English, the possessive
that is sleeping, constitutes a referential unit. Tomasello reasons that
marker and the noun, –‘s book, must undergo pied piping, resulting
children appear to conform to a structure-dependent rule because
in the complex wh-phrase, whose book. English does not tolerate
they are disinclined to interrupt functional units that are used to
whose-questions such as Whose did he buy book? or Who do you
perform acts of referring.
think’s book did he buy?
Pied-piping is obligatory in English whose-questions, but it
2.2. Displacement in human languages is not obligatory in many languages (e.g., Hungarian, Chamorro,
Slavic languages). These languages permit the displacement of a
There may be some inherent plausibility in this account of wh-possessor word from the remainder of the phrase yielding
children’s behaviour in producing Yes/No questions with rela- whose-questions like (8) and (9), from Hungarian. Example (8)
tive clauses, such as (5). It is unclear whether Tomasello (2008) illustrates the (optional) extraction of the wh-possessor word ki-
intended this pragmatic account of the absence of structure- nek ‘who-Dative’ from the remainder of the possessive phrase in
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 125

a matrix question, and (9) illustrates the (optional) extraction of like Is mummy beautiful?, a recurrent network ‘learns’ to produce
the wh-Possessor ki-nek ‘who-Dative’ from the remainder of the adult-like questions corresponding to declarative sentences with a
possessive phrase in a long-distance question. relative clause, because the network downgrades the probability of
‘smoking’ (or any progressive verb form) following ‘who’. The claim
by Ambridge et al. is that children avoid structure-dependence
errors because they are distributional analyzers.
If children’s production of adult-like Yes/No questions is the
result of their replication of the co-occurrence patterns they
encounter in the input, then this leads to clear empirical predic-
tions. For example, children would be expected to make more
structure-independent errors in cases where the transitional prob-
ability of a given bigram is higher than that of the bigram ‘who
smoking.’
The Continuity Assumption allows for the possibility that Now consider the declarative sentence (12). This sentence con-
English-speaking children break up functional units in whose- tains the singular noun, boy, followed by the relative pronoun, who,
questions, as in Hungarian. If so, the grammars of children acquiring the modal auxiliary verb, can, and the uninflected verb run. This
English would generate whose-questions that are unacceptable for brings together the well-formed substring boy who can run. Let us
adults. This is exactly what was found in an elicited production see what extraction of the auxiliary means in terms of the probabil-
experiment reported in Gavruseva and Thornton (2001). Several of ity of resulting bigrams. If we extract can out of the relative clause
the child participants in this study produced split English whose- boy who can run, we end up with the string boy who run. Even though
questions, such as (10). this trigram has low probability, the bigram who run that is con-
tained in it, has a high probability, as it is completely acceptable
in sentences like Boys who run are usually fit. That is, the bigram
who run has a higher probability than the bigram who smoking.
According to Ambridge et al., then, this means that children should
In (10), the adult wh-phrase whose book is divided into two parts, produce structure-dependence errors such as (13) more often than
with the wh-word who separated from the remainder of the posses- ones like (11).
sive phrase, ‘s book. For adults, the phrase whose book must remain
intact, yielding Whose book do you think is on the table? The fact that
English-speaking children break apart whose-phrases runs counter
to Tomassello’s proposal that children keep functional units intact.
On the biolinguistic approach, there is no constraint, pragmatic or
otherwise, that compels strings of words to “stay together as one
Notice what happens, in addition, when the plural noun,
functional unit.” If, as Tomasello argues, such a constraint prevents
boys, is substituted for the singular noun, boy. Example (14) is
children from extracting an auxiliary verb from a relative clause
a declarative sentence with the plural noun ‘boys.’ The illicit
in Yes/No questions, then it is surprising that the constraint is not
structure-independent Yes/No question corresponding to this
more widespread in human languages, and that children and adults
declarative sentence is (15), which contains, apart from the bigram
consistently disregard it in so many linguistic structures.
who run, the higher frequency trigram boys who run. The trigram
The biolinguistic approach offers an alternative account of the
boys who run is a well-formed functional unit, in contrast to the tri-
absence of Yes/No questions in which an auxiliary verb is extracted
gram boy who run. So the trigram boys who run is much more likely
from inside a relative clause, such as the unacceptable Yes/No
to be in the input to children than boy who run or boy who smoking.
question, Is the boy who smoking is crazy? The account takes the
On the distributional account of children’s structure-independent
form of a domain specific, cross-linguistic constraint that prevents
errors, therefore, children are expected to make an even greater
displacement of linguistic material that resides in a certain struc-
number of structure-independent Yes/No questions with plural
tural position in sentence structures. The structure dependence of
nouns that are modified by relative clauses, as compared to singular
children’s Yes/No questions is just one instance of this general con-
nouns (examples are adapted from Ambridge et al., 2008).
straint on the extraction of linguistic material (see Berwick et al.,
2012; Crain and Pietroski, 2001).

2.3. Children as distribution analyzers

There is another proposal for why children do not make (or make Ambridge et al. (2008) tested these predictions in two experi-
very few) structure-independent errors. Ambridge et al. (2008) ments based on the methodology used in the Crain and Nakayama
propose that children do not make structure-independent errors study. Children were presented with declarative sentences with
such as (11) because the bigram ‘who smoking’ is infrequent in the both singular nouns and ones with plural nouns, either with aux-
input that children experience. As distributional analyzers, children iliary verb, can, or with a form of the copula, be. The task for the
are unlikely to produce utterances that group infrequent words child participants was to convert these declarative sentences into
together. Yes/No Questions. In contrast to the Crain and Nakayama study, the
child participants in the Ambridge et al. study did produce some
non-adult Yes/No questions like (11), (13) and (15), in which the
auxiliary verb was absent from a relative clause. However, these
According to Ambridge et al. (2008), children’s production of errors were infrequent. The study found no significant difference in
questions involves the same procedure as a recurrent network per- error rates between the different types of Yes/No questions. Non-
forming a word prediction task. Based on training with simple adult Yes/No questions with singular nouns constituted 7% of all
declarative sentences such as Mummy is beautiful and questions
126 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

scorable2 responses, and non-adult Yes/No questions with plural the main clause, as a modifier of the verb swim. It follows that
nouns constituted 9% of children’s responses. The error rate for the declarative counterpart to the Yes/No question is (16a), not
questions with a form of the copula verb, be, did not differ sig- (16b). Native speakers of English not only have knowledge about
nificantly from the error-rate for questions with the auxiliary verb, the well-formedness of surface strings, but they also know which
can. Based on these null findings, it seems unlikely that the acquisi- interpretations are compatible with a particular surface string and
tion of Yes/No questions involves a straightforward application of which are not. Moreover, there is no way to invoke either func-
the same mechanisms that are used in a recurrent network. tional units of reference, or distributional analysis to explain why
In summary, less than 10% of children’s responses in the native speakers know that the auxiliary verb can has been extracted
Ambridge et al. study were non-adult-like. Ambridge et al. from the main clause in (16), and not from the relative clause. The
conclude that their findings falsify the claim that children string eagles that fly is as much a functional unit as eagles that
never postulate structure-independent operations: “Thus whilst swim, and it is highly unlikely that adding the auxiliary verb can
structure-dependence errors are by no means frequent, they would to either string results in a different probability of occurrence in
seem to be made by a reasonably high proportion of children, at children’s experience, i.e., eagles that can fly versus eagles that can
least for questions with the modal auxiliary CAN” (p. 233). How- swim. As such, the usage-based proposals based on functional units
ever, there is no evidence that a structure-independent operation or distributional analyses seem ill-equipped to handle our native
was the source of any child’s responses. Non-adult responses were speaker intuitions about sentences like (16). These semantic intu-
not characteristic of individual children, but were spread among the itions about the underlying structure of Yes/No questions follow
child participants. This suggests that children’s non-adult produc- from structure-dependence, and are backed up by facts about mor-
tions were errors in performance, and not the result of a structure phological agreement, such as those exhibited in (17) and (18),
independent operation. which are taken from Chomsky (2012).

2.4. Structure-dependent interpretations

The distributional analysis of children’s Yes/No questions also

fails to engage with the interpretations that children and adults This is not just an isolated case at far remove from children’s
assign to sentences. Consider the often-cited Yes/No question (16). experience. Native speakers of English also know immediately that
This question contains two verbs (fly and swim) and the sentence- the Yes/No question in (19) is derived from the declarative sentence
initial modal auxiliary verb can. From a logical point of view, it is (19a), and not from (19b). Again, these native speaker intuitions
unclear where the auxiliary verb can originated. The two options reflect our knowledge that Yes/No questions cannot result from a
are presented in (16a,b). structure-independent operation. And again, these intuitions about
interpretation cannot be explained by appealing to functional units
of information or to differences in distributional frequency. (Exam-
ple (19) was adapted from Berwick et al., 2011).

Chomsky has introduced numerous arguments showing that the

initial auxiliary verb is associated with the verb swim, and not with
closer verb fly. Chomsky makes the relevant points in the following
passage (2013, p. 651–652). The evidence that human languages do not use local well-
formedness as the basis for sentence meaning is not limited to
“Consider the sentence ‘instinctively, eagles that fly swim.’ The
abstract cases taken from adult speech. An experiment conducted
adverb “instinctively” is associated with a verb, but it is swim,
by Gualmini and Crain (2005) showed that 3- to 6-year-old English-
not fly. There is no problem with the thought that eagles that
speaking children do not group sequences of words into semantic
instinctively fly swim, but it cannot be expressed this way. Sim-
units based on local well-formedness relations. The experiment
ilarly the question “can eagles that fly swim” is about the ability
presented sentences like (20) to children. This test sentence con-
to swim, not to fly. What is puzzling about this is that the asso-
tained the well-formed substring ‘he cannot lift the honey or the
ciation of the clause-internal elements “instinctively” or “can”
doughnut’ (21).
to the verb is remote and based on structural properties rather
than proximal and based on solely linear properties, a far sim-
pler computational operation and one that would be optimal
for processing language. . . . In technical terms, the rules are
invariably structure-dependent, ignoring linear order. The puzzle
is why this should be so − not just for English but for every lan-
guage, not just for these constructions but for all others as well,
over a wide range. . . . What is it about the genetically deter-
mined character of language − UG − that imposes this particular Notice that the substring yields the ‘neither’ interpretation of the
condition?” disjunction word ‘or’. This interpretation is assigned because, in the
substring indicated in (21), the disjunction word ‘or’ appears inside
As this passage from Chomsky makes clear, our native speaker
the scope of negation. However, the ‘neither’ interpretation of the
intuitions tell us that the auxiliary verb can in (16) originated in
disjunction word ‘or’ is not available if language-users assign a hier-
archical structure to sentence (20). Then, the substring in (21) is
part of a relative clause, as indicated in the structural representation
2
Scorable responses are a proper subset of children’s responses. Looking at all of in (22).
children’s responses would reduce the percentages of structure-dependence errors,
e.g., to around 5% for the plural and singular questions with ‘can’ combined.
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 127

According to the usage-based approach, by contrast, there are no

such structural constraints either in child or adult language. Instead,
the usage-based approach contends that “the facts attributed to
the binding principles reduce to a very simple functional explana-
In the hierarchical representation, negation resides inside the
tion” (Ambridge et al., 2014; p. e80). According to the usage-based
relative clause and, therefore, cannot take scope over the disjunc-
approach, children learn information-theoretic principles (e.g.,
tion word ‘or.’ The interpretation of (20) that results is the ‘not
noun phrase accessibility) that “could replace the need for innate
both’ meaning, rather than the ‘neither’ interpretation that would
syntactic constraints” (Matthews et al., 2009; p. 605). One mecha-
be assigned if the sentence was treated as a string of words.
nism that plays an important role in anaphoric relations according
Gualmini and Crain (2005) tested children’s interpretation of
to the usage-based approach is pre-emption. For example, pre-
the disjunction word ‘or’ in sentences like (20). Children correctly
emption prevents children from assigning co-reference between
accepted sentence (20) 80% of the time as a description of a story in
a pronoun and a referential noun phrase if both appear in the
which Karate Man had given one of the Pooh Bears a doughnut, but
same simple sentence. An example is Joe adores him. According to
not the honey. This shows that children do not analyse negation as
Matthews et al. (2009, p. 605) “co-reference in sentences such as
taking scope over the disjunction word ‘or’ in the test sentences.
‘Joe adores him’ is not so much ruled out as pre-empted by sen-
If so, children would have rejected the test sentences. This finding
tences like “Joe adores himself”.”3 ; Likewise, Boyd and Goldberg
indicates that children assigned an adult-like hierarchical structure
(2011, p. 55) contend that pre-emption is the means by which
to sentence (20), as represented in (22). Children did not assign an
“speakers learn not to use a formulation if an alternative formula-
interpretation consistent with the substring in (21).
tion with the same function is consistently witnessed.” Essentially,
This example and numerous others reveal the empirical inade-
the idea is that children repeatedly encounter sentences with a
quacy of the usage-based approach. According to one advocate of
reflexive pronoun such as Joe adores himself in circumstances that
the usage-based approach (Lieven, 2010; p. 2547) “children build
depict a reflexive event. Pre-emption leads children to refrain from
their grammars initially out of the phonological–lexical strings that
using ordinary pronouns, such as him, in these same circumstances.
they learn from the input rather than analysing that input in terms
Before long, children use the pronoun him and the reflexive pro-
of pregiven, more abstract, linguistic categories.” Instead of see-
noun himself in the same way as adults do, restricting the use
ing children as projecting hierarchical structure in order to derive
of himself to circumstances in which the referent of the Subject
adult-like interpretations for sentences such as (21), the usage-
noun phrase stands in some abstract relation to himself, e.g., he
based approach contends that children are sensitive to “surface
adores himself, or performs some action upon himself. Thereafter,
co-occurrence patterns in the input data” (Ambridge et al., 2008;
the ‘reflexive’ meaning is reserved for sentences with reflexive
p. 234). If so, then nothing would prevent children from assigning
pronouns such as himself, and is not assigned in sentences with ordi-
the ‘neither’ interpretation to the disjunction operator in the sub-
nary pronouns such as him. At that point, child language matches
string (21), but children never make this assignment, as Gualmini
the adult language in this respect.
and Crain showed.
There is an immediate problem with this account, however. An
To conclude this section, the alternatives proposed by the usage-
extensive literature on this topic reveals some overlap in the use of
based account for the absence of structure-independent errors are
ordinary pronouns and reflexive pronouns. Examples from English
not convincing. On the one hand, the domain general mechanisms
are provided in (23). The overlap in interpretations is indicated by
invoked by the usage-based approach to language acquisition are
the indices on the NPs. We will adopt the usual conventions: NPs
too strong, because they predict errors that children do not make.
that have the same index are interpreted as picking out the same
On the other hand, these mechanisms are too weak, because they
referents(s); they are said to be anaphorically related, or corefer-
fail to account for the possible interpretations that children do and
ential. NPs with different indices are interpreted as picking out
do not assign to sentences.
different individuals, and are not anaphorically related; they are
said to be disjoint in reference, or non-coreferential. (This example
is from Reinhart and Reuland 1993, p. 661).

3. Anaphoric relations

From a biolinguistic perspective, the initial state in language

acquisition consists of the same structure-dependent linguistic
principles that govern adult languages. We saw how such prin-
ciples operated in both adult and child language in the previous In the case of the sentences in (23), the indices indicate that the
section, where we discussed how English Yes/No questions were ordinary pronouns, him and her, may be anaphorically related to the
derived from their declarative sentence counterparts. Subject NP. This is surprising on the usage-based account, because
This section takes up another class of structure-dependent phe- the examples in (23) show that reflexive pronouns, himself and
nomena, ones that restrict the interpretations of noun phrases of herself, are also permitted. Presumably, pre-emption should pro-
various kinds, including ordinary pronouns such as he and him, hibit coreference between the ordinary pronouns and the referring
reflexive pronouns like himself, and referring expressions such as expressions. The fact that ordinary pronouns are acceptable under
Papa Bear. Anaphoric relationships between noun phrases are per- coreference shows that other factors are at work in such examples.
mitted in certain structural configurations, but not in others. The Moreover, if children were to use a pre-emption strategy to learn
constraints on anaphoric relations are stated in a series of binding constraints on anaphoric dependencies, they would never master
principles (e.g. Chomsky, 1981; Reuland, 2011). The biolinguistic cases like (23). That is, children who applied pre-emption would not
approach anticipates that child language learners will adhere to converge on a grammar that is equivalent to that of adult speakers
the same structural constraints that are exhibited in adult lan- of English. We turn next to the more far-reaching question – the
guages, including the binding principles. These binding principles,
and other linguistic constraints, are seen to be part of the innately
specified Universal Grammar, so children are expected to exhibit 3
See Yang (2015) for discussion of several formal problems in an account of pre-
knowledge of anaphoric relations as soon as they can be tested. emption based on statistical analysis.
128 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

extent to which information-theoretic principles are the equals of noun phrase, John’s mother, originated in object position, following
structural principles. the verb adores. This explains why the declarative sentence John’s
mother adores dearly is unacceptable; it lacks a Direct Object. Top-
3.1. Blocking coreference icalization takes the topic phrase John’s mother, and positions it at
the front of the sentence. However, it moves back into its origi-
From a biolinguistic perspective, sentences are hierarchically nal position at the level of semantic interpretation. This process of
structured, and linguistic principles (e.g., the binding principles) interpreting an expression twice, once in its surface position, and
constrain the interpretations that can be assigned to them (e.g., a second time at a different position, is called reconstruction. The
Everaert et al., 2015). In contrast, the usage-based approach con- process of reconstruction for (25) is schematically depicted in (26)
tends that the adult final-state of language acquisition does not and graphically depicted in Fig. 3.
constitute a generative grammar, but, rather, a set of constructions
(Ambridge and Lieven, 2011; p. 123). Structure-dependent rules
therefore have no role on the usage-based approach. Instead of
structure-dependent principles, the usage-based approach invokes
information-theoretical principles to account for the same linguis-
tic phenomena. For example, the interpretation of pronouns in
After reconstruction, the semantic representation of sentence
single clause sentences is subject to the following principle: “if
(25) is identical to that of sentence (24). In both cases, corefer-
a pronoun is used as the topic, this indicates that the referent is
ence between the pronoun he and the lexical NP John is ruled out.
highly accessible, rendering anomalous the use of a full NP any-
More specifically, coreference is ruled out in both (24) and (25)
where within the same clause” (Ambridge et al., 2014; p. e77, fn.17).
because John resides in the structural domain of the pronoun he
To unpack this principle we need to know how to identify the topic
(i.e., following reconstruction in the case of (25)).
of a sentence. On the usage-based approach, “the topic/theme is the
Technically speaking, the structural relationship between the
NP that the sentence is ‘about’, and about which some assertion is
pronoun and the referring expression is known as c-command. It
made (the comment/focus/rheme)” (Ambridge et al., 2014; p. e77).
may be useful to think of c-command as sentence scope, so if a noun
We will examine this proposal as it pertains to some of the
phrase A c-commands another noun phrase B, then A takes scope
examples discussed in Ambridge et al. (2014). First, these usage-
over B. One of the binding principles (called Principle C) dictates
based researchers say that the principle just expounded explains
that coreference is ruled out whenever a pronoun takes scope over
why example (24) is unacceptable. In example (24), the pronoun,
a referring expression. As a result of reconstruction, the pronoun he
he, is the topic of the clause. As a result of the discourse principle
in the Topicalized sentence in (25), John’s mother, he adores, takes
under consideration, the lexical NP John cannot be used to refer
scope over John, just as it does in the declarative sentence (24).
to the same person as the pronoun does. To introduce some ter-
Hence, coreference is ruled out in both (24) and (25).
minology, when a pronoun precedes a lexical NP, as in (24), this
By contrast to this unified account of linguistic phenomena,
is referred to as backward anaphora, and when a lexical NP pre-
the usage-based approach treats each construction as unique. The
cedes a pronoun, this is called forward anaphora. As the examples
usage-based approach lacks structural constraints on anaphoric
in (24) indicate, coreference between a pronoun and a lexical NP is
dependencies, and it also lacks mechanisms that displace (move)
tolerated in sentences with forward anaphora, but coreference is
or reconstruct phrases from one position in a structural represen-
blocked in certain cases of backward anaphora.
tation to another position. On this approach, therefore, there is no
straightforward way to rule out both (24) and (25), as it treats the
two sentences as unrelated linguistic units. The problem for the
usage-based approach is that no single information-theoretic prin-
ciple could apply to both sentences. In example (25) the sentence
is clearly about John’s mother, and not about John. So, the phrase
Ambridge et al. (2014) propose to account for the unacceptabil- John’s mother is the topic of sentence (25). By contrast, the pronoun,
ity of (25) in the same way. We question this account, however, as he, is the topic of sentence (24). Because John’s mother and not the
we now discuss. pronoun he is the topic of (25), the sentence does not violate the
discourse principle proposed by Ambridge et al. Given that the dis-
course principle is inoperative in (25), the pronoun he should be free
to refer to the same individual as the lexical NP, John, just as it does
in other cases of forward anaphora, e.g., John’s mother likes his tie.
Sentence (25) is an example of a syntactic process called Top- In short, the account proposed by the usage-based approach incor-
icalisation. In the Topicalized structure (25), the sentence-initial rectly predicts that (25) should be acceptable on an interpretation
that takes John’s mother and he to be coreferential.4
In response, Ambridge et al. might contend that the Subject noun
phrase is always the topic of the sentence. If so, example (25) would
be ruled out, because he would be the topic in (25), just as it is in
(24). However, this explanation of non-coreference turns on the
structural notion of Subject, rather than the information-structural
notion of topic. If the operative principle is about Subjects rather
than topics, then the usage-based approach runs into further diffi-
culties. One difficulty would be to account for the acceptability of

4
The same holds for the other examples of Topicalisation in their footnote 17.
Still further cases can be found in the literature, such as (i). (i) *Proud of Johni , hei
Fig. 3. Reconstruction of the topic phrase ‘John’s mother’. was. (Heycock and Kroch, 2002) (71a)).
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 129

(27) and (28), as compared to the unacceptability of (29). (Example 3.2. Thornton et al. (2015)
(28) was adapted from Sportiche, 2006).
The prediction of the biolinguistic approach was empirically
investigated by Thornton et al. (2015) using cleft sentences like
(30). The study presented test sentences such as (32) and (33)
to preschool children, as well as the control sentences indicated
beneath them.

In (27), the pronoun he is the Subject/topic. This means,

according to the information-theoretic principle, that coreference
between Ian and he should be out. In contrast to this prediction
however, coreference with Ian is allowed. Similarly, the usage-
based approach incorrectly predicts that the pronoun he and the
referring expression John cannot be coreferential in (28). Notice, Adults interpret the test sentence in (32) to require disjoint ref-
also that, because the pronoun, him, resides in the object position erence between the pronoun he and the name Spot, just as in the
in (29), it cannot be the Subject or the topic, so coreference should control sentence, He brushed Spot, where the pronoun precedes the
not be ruled out according to the information-theoretic principle. name in the surface syntax. In the test sentence in (32), the name
Again, this is contrary to fact. Spot undergoes reconstruction at the level of semantic interpreta-
The account offered by the usage-based approach for the assign- tion. Coreference is prohibited in both the test sentence in (32) and
ment of anaphoric relations in multi-clause sentences is similar to in the control sentence, because the subject NP, Spot, is in the struc-
the account of single clauses they offer, as the following remark tural (c-command) domain of the pronoun at the level of semantic
makes clear: “In general, it makes pragmatic sense to use a lexi- interpretation. This is depicted in (34); the reconstructed Subject
cal NP (including quantified NPs like everyone) as the topic about NP Spot’ is inside the angled brackets. As advocates of the usage-
which some assertion is made, and a pronoun in a part of the sen- based approach point out, in general pronouns can be anaphorically
tence containing information that is secondary to that assertion, linked back to ‘full’ noun phrases that precede them. However, the
but not vice versa.” There is an additional principle at work in sen- pronoun cannot be anaphorically linked back to the full NP that
tences with multiple clauses. The principle is this: “once a speaker precedes it in (32).
has already referred to an individual with a full NP, it is quite natu-
ral to use a pronoun in a subsequent clause, and indeed, unnatural
not to [. . .].” It follows that the only cases in which coreference is
ruled out are sentences in which a pronoun which appears in the
topic position precedes a noun phrase that appears in a clause that
Reconstruction also applies to the cleft sentence (33), as indi-
expresses presupposed information, i.e., a backgrounded clause. On
cated in (35). In this case, reconstruction of the noun phrase enables
this account, however, sentences like (30) and (31) are predicted to
the assignment of a bound pronoun interpretation of the noun
be acceptable, again contrary to fact. The NP precedes the pronoun
phrase, her pig. Therefore, example (33) licenses the meaning that
in each of these examples, so coreference should be possible, but it
every girl carried her own pig in addition to the meaning on which
is not.
every girl carried some other girl’s pig. The surface form of the
sentence provides no clue to the ambiguity. The bound pronoun
interpretation is licensed under c-command due to reconstruction.
Since the surface forms in cleft sentences such as those in (32) are
not indicative of the prohibition on interpretation that is enforced
by adults, a usage-based approach would presumably not predict
that young children assign the same constraint on interpretation
The examples in (30) and (31) are cases of reconstruction, as adults do to such sentences. This is what the generative account
as in example (25). Because the usage-based account is based predicts, however.
on strings, rather than structures, it has no recourse to syntac- The Thornton et al. study of children’s comprehension of cleft
tic operations (e.g., reconstruction) that reposition noun phrases sentences and the control sentences used a research methodology
from one place to another. But this prevents the usage-based called the Truth Value Judgment. In this task, stories are acted out
approach from using syntactic structure to determine whether or in front of the child participants in the experimental workspace
not an anaphoric dependency is allowed. Sentences that involve using toy characters and props. While one experimenter acts out
the reconstruction of constituents will consistently run counter to the stories, a second experimenter plays the role of a puppet. At the
the information-theoretical mechanisms postulated by the usage- end of each story, the puppet produces one of the test sentences,
based approach. On the other hand, the facts about coreference and or a control sentence. The child’s task is to say whether the puppet
non-coreference follow naturally from the biolinguistic approach. “said the right thing”, i.e., whether or not the puppet’s statement
Again we would emphasize that the arguments we are making are was a true or false description of the events that took place in the
not about idiosyncratic facts about adult English. Not only does the story. If the child judges that the puppet said something true, then
biolinguistic approach predict that the constraint on interpretation it is assumed that the child’s grammar generates a structure and
that is witnessed in sentences like (30) and (31) should be known meaning for the sentence that matches the events that took place
to English-speaking children as soon as they can be tested, but this in the story. If the child judges the puppet’s statement to be false,
approach predicts that children across the globe should adhere to this is taken as an indication that the child’s grammar generates a
this constraint on interpretation as it applies to a range of linguistic structure and meaning that does not match the events in the story.
structures. We return to this point in Section 7. A corollary assumption is that, whenever possible, children (and
adults) access a meaning that makes the puppet’s sentence true.
130 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

This is called the Principle of Charity. Therefore, when the child differs from the assignment of the same discourse referent to two
judges the puppet’s statement to be false, it is inferred that the linguistic expressions (e.g. Reuland, 2001).
child was unable to access any sentence-meaning pair that made
the puppet’s statement true. That is, children’s rejections of the
puppet’s statements are evidence that the sentence is unambiguous
for the child, and evidence that the only interpretation permitted The difference between quantified and referential NPs has
by the child’s grammar is one that does not match the story con- proven to be important in child language research. A well-known
text. Children’s consistent rejections of test sentences in the Truth observation from the acquisition literature is that children as old
Value Judgment task provide evidence that their grammars impose as 6;6 assign a non-adult interpretation to sentences with referen-
a constraint on interpretation, for example structural constraints tial NPs combined with ordinary pronouns in object position. An
on anaphoric dependencies. Children who are too young to suc- example is the sentence Papa Bear is washing him. For children, this
cessfully perform other psychological tasks, such as judgments of sentence is true in two circumstances, one in which Papa Bear is
grammaticality, have proven able to produce reliable judgments of washing some other male individual, and one in which Papa Bear
truth and falsity using the Truth Value Judgment task. is washing himself. Adults only judge the sentence to be true in the
Twenty children participated in the Thornton et al. study, as first of these circumstances. Interestingly, at the same time, chil-
well as a control group of 20 undergraduate students. The chil- dren do not misconstrue sentences with reflexive pronouns, such
dren ranged in age from 4;0 to 5;5, with a mean age of 4;9 years. as Papa Bear is washing himself. Children display adult-like judg-
Here are the main findings. The child participants rejected coref- ments in response to sentences with reflexive pronouns by age 4
erence in the cleft sentences such as (32) 94% of the time for the (Chien and Wexler, 1990; Jakubowicz, 1984; Deutsch et al., 1986
clefts, and they rejected coreference in the control sentences 99% on English; Koster, 1993; Philip and Coopmans, 1996 on Dutch).
of the time. The twenty adult participants rejected both the tar- Children’s delay in assigning the adult-like meaning to sen-
get and the control sentences 100% of the time. Turning to bound tences like Papa Bear is washing him is unexpected both on
pronoun cleft sentences such as (33), children accepted the bound the usage-based approach, and on the biolinguistic approach
pronoun interpretation 65% of the time, whereas they ruled out (Matthews et al., 2009). Advocates of the usage-based approach
this interpretation of the control sentences 83% of the time. The conjecture that children’s limited experience with such sentences
adult participants produced a similar pattern of responses. They may be responsible, at least in part, for the delay they experience
accepted the bound pronoun interpretation in the cleft sentences, (recall that a reflexive interpretation of pronouns in cases like Papa
as in (33), 50% of the time, but rejected this interpretation of the Bear is washing him is not ruled out, but pre-empted according
control sentences 83% of the time. In view of the ambiguity of cleft to the usage-based approach). That is, if the child has not heard
sentences such as (33), it is not surprising that both children and sufficient instances of sentences with a reflexive pronoun, then
adults sometimes chose the alternative, direct reference, interpre- the coreference interpretation of ordinary pronouns might not
tation, according to which the pronoun her functions as a deictic be blocked, because pre-emption will not be sufficiently strong.
pronoun in the expression her pig. Children had no difficulty inter- Children’s performance is therefore expected to gradually improve
preting binding relations in sentences requiring reconstruction, with experience.
despite the fact that the surface form and, hence the adult linguistic From a biolinguistic perspective, the structural principles on
input, is uninformative about the impossibility of assigning certain anaphoric dependencies are expected to be in place from the ear-
interpretations to cleft structures. liest stages of acquisition. It has been proposed, therefore, that the
difficulties children experience in mastering the non-coreference
facts in sentences with referential NPs and pronouns is not due to a
3.2. Language delay: illicit coreferences lack of syntactic knowledge. Rather, children are seen to experience
difficulty in executing certain pragmatic principles that govern
In the previous section we found that the information-theoretic the interpretation of pronouns, potentially because they lack the
principles used by the usage-based approach do not suffice in required processing resources (see Grodzinsky and Reinhart, 1993;
accounting for children’s rapid acquisition of knowledge about the Thornton and Wexler, 1999). If this account for children’s non-adult
anaphoric relations that can and cannot hold between different responses is on the right track, then children are expected to per-
kinds of noun phrases. This section examines a case where child form without fail in responding to sentences that are not governed
and adult language differs in the assignment of anaphoric relations. by pragmatic principles, but are instead governed by structural
To understand the phenomenon, we must first make a distinc- principles. The sentences in question involve replacing a referen-
tion between binding and coreference. A binding relationship is tial noun phrase, e.g., Papa Bear, by a quantificational noun phrase,
one in which one expression is dependent on another for its inter- such as every bear, as in (37). As we have seen, according to linguis-
pretation. By contrast, coreference is a relationship in which two tic theory, quantificational NPs such as every bear do not permit a
expressions pick out the same referent in a domain of discourse. coreferential relation with a pronoun, because they do not refer.
Coreference is witnessed in sentences with referential noun Quantificational NPs can only bind pronouns, rather than estab-
phrases, such as John thinks he is clever, where the pronoun he can lishing a coreferential relation with them, as referential NPs do.
pick out the same individual as the name John.5 An example of Therefore, the anaphoric relations between quantificational NPs
binding is given in (36). Intuitively, it is clear that there is no ref- and pronouns are regulated by structural principles, and not by
erent corresponding to the quantificational expression no one in pragmatic principles.
(36). Nevertheless, the pronoun he can depend on the quantifica-
tional expression no one for its interpretation, when the pronoun
is not being used deictically to refer to someone in the domain of
discourse. This dependency is linguistically encoded and crucially

The situational contexts corresponding to the sentences in (37)

and (38) in the experiments were clearly different, because there
5
Note that pronoun he in this particular structural configuration can also be bound
was a salient single individual, Papa Bear, in the context corre-
by John. The bound variable representation yields the same meaning representation
as the coreference representation. sponding to (38), but there were several bears in the context
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 131

corresponding to (37). It has been suggested that such differences the usage-based approach to language acquisition, but it is entirely
in context could have influenced children’s interpretation of the consistent with the biolinguistic approach.
pronouns in the test sentences (see, e.g., Elbourne 2005; Conroy
et al., 2009). However, one study of the contrast between quantifi- 3.3. Guasti and Chierchia (2000)
cational NPs and referential NPs, by Thornton (1990), was immune
to this criticism. The Thornton study compared indirect questions At this point, we have discussed reconstruction effects in child
and their answers, such as (39), and declarative sentences with plu- language, and we have discussed the distinction between bind-
ral referential NPs, such as (40). The wh-question Who scratched ing and coreference. A study reported in Guasti and Chierchia
them? was expected to pattern in the same way as a sentence with (2000) investigated Italian-speaking children’s understanding of
a quantificational NP, such as Every bear is washing him. sentences involving reconstruction to assess their knowledge that
anaphoric binding is determined by the structural properties of
sentences, rather than by discourse principles, as proposed by
Ambridge et al. (2014). Guasti and Chierchia (2000) investigated
children’s understanding of sentences where a prepositional phrase
(PP) is fronted to sentence-initial position. Example (41) illustrates
the phenomenon, which is called PP-preposing.
In the Thornton study, test sentences like (39) and (40) were pre-
sented to children in identical contexts. Nevertheless, the twelve 3-
to 4-year-old child participants (who ranged in age from 3;7 to 4;8)
accepted coreference in (40) on half of the trials, but the same chil-
dren consistently (92%) rejected sentences like (39) in the same
context. That is, on half of the trials, children interpreted sentence The semantic representation associated with (41) is depicted
(40) to mean that Ernie and Bert scratched themselves. However, in (42). As indicated in (42), the preposed-PP is reconstructed to
the same children assigned this same interpretation to sentences its original position at the level of semantic interpretation. Fol-
like (39) on fewer than 10% of the trials. This is additional confir- lowing reconstruction, the referring expression, John, is positioned
mation for the conclusion that children’s non-adult interpretations inside the structural domain of the pronoun, he. Therefore, the
of sentences with referential NPs are due to a delay in the acquisi- interpretation that is assigned to (41) analyzes the pronoun and
tion of pragmatic principles, and did not represent violations of the the referring expression to be disjoint in reference, as in the cor-
structural binding principle. responding declarative sentences such as (43), where the PP near
As such contrasts make clear, the alternative approaches to lan- John’s bicycle appears in situ in the sentence structure.
guage acquisition make different predictions about the patterns
of children’s linguistic behaviour that are expected to be mani-
fested in the course of acquisition. From a biolinguistic perspective,
structural principles are available to children from the outset of Despite PP-preposing, therefore, both sentences (41) and (43)
language development, whereas children need time to expand are assigned the same interpretation, with coreference ruled out
their processing resources and/or to develop pragmatic skills.6 in both sentences by a structural constraint (i.e., Principle C of the
For the usage-based approach, input frequency plays the major binding theory). This would not be expected on any account that
role in determining the pattern of acquisition. On the biolinguistic is based on the surface string of words. Sentences where the pro-
approach, therefore, children should not display difficulty in inter- noun appears first are referred to as backward anaphora, and ones
preting the pronoun in sentences with quantificational NPs, such as in which the referring expression appears first are called forward
(37) and (39), even at the stage at which they experience difficulties anaphora. As Guasti and Chierchia remark (2000, pp. 130–131):
in responding to sentences with referential NPs, such as (38) and
“According to reconstruction approaches, the structures of [our
(40). This prediction about the contrast between these sentences
examples 41 and 43] are virtually identical and, hence, the rul-
does not follow from the usage-based approach. The usage-based
ing out of the two cases has to be uniform. If Universal Grammar
approach predicts no differences in children’s responses to sen-
drives acquisition, this seems to predict that as soon as children
tences with referential NPs or with quantificational NPs as “only
get (43) right, they must also get (41) right. This is the oppo-
generativist approaches to language acquisition predict that chil-
site of what one would expect just by considering how forward
dren will have knowledge of the relevant syntactic constraint”
and backward anaphora work in general. Thus, looking at recon-
(Matthews et al., 2009; p. 607).
struction effects in child grammar might help us choose among
The difference in children’s pattern of responses to sentences
these different hypotheses.”
with referential NPs and ones with quantificational NPs has been
documented extensively in the literature, and we will not review Returning to the pragmatic account proposed by Ambridge et al.
the evidence here. Suffice it to say that there is ample evidence (2014), the PP-preposing example in (41) (repeated here) is on a
that children perform significantly better on sentences with quan- par with (44), rather than being on a par with (43). Since the pre-
tified NPs than ones with referential NPs, and the difference is found posed lexical PP, near John’s bicycle, is the topic (what the sentence
across languages (see e.g., Chien and Wexler, 1990; Matthews et al., is about), it should be quite natural to use a pronoun to refer to the
20097 ; Philip and Coopmans, 1996). This result is unexpected on same individual, as (44) illustrates.

6
Of course, the child might still not display at-ceiling performance in particular
experimental tasks as a result of task demands (Crain and Thornton, 1998).
7
The children in the Matthews et al. study did not perform at ceiling in responding
to sentences with quantified NPs, in contrast to earlier studies. It is worth asking According to the pragmatic account, therefore, (41) should be
why the Matthews et al. study obtained different results. But the main point is that
children did perform significantly better in responding to sentences with quantified
accepted by children (and adults in fact) in the same circumstances
NPs than to ones with referential NPs in the Matthews et al. study. This difference that validate (44), when John saw a snake near his own bicycle.
is not expected on the usage-based account. Again, the biolinguistic approach and the usage-based approach
132 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

make opposite predictions about how sentences with PP-Preposing (Kirjavainen and Theakston, 2009). For example, Tomasello (2000,
will be interpreted. The usage-based account predicts that corefer- p. 71) remarks:
ence will be permitted by children for sentences like (41), whereas
“. . . children hear things like ‘Let her open it’ or “Help her open
the biolinguistic account predicts that coreference will be prohib-
it” all the time, and so it is possible that when they say these
ited by children.
things they are simply reproducing the end part of the utter-
The Guasti and Chierchia study introduced another innovation.
ances they have heard.”
We saw in Section 3.2 that children respond differently to sentences
with quantificational expressions (e.g., every bear) as compared And Kirjavainen and Theakston (2009, p. 1094) address the fact
to sentences with referential expressions (e.g., Papa bear). Chil- that such non-adult uses of pronouns are optional as follows:
dren were found to reject anaphoric relations significantly more
“Within this approach, children are expected to extract lexically
often when coreference is unavailable (e.g., Every bear washed him),
specific chunks from complex but relatively frequent utterances
rather than when coreference is allowed (e.g., Papa bear washed
in the input, as well as learning shorter utterances as a whole.
him). Based on this asymmetry, Guasti and Chierchia investigated
Thus, errors where a NOM pronoun is erroneously replaced
Italian-speaking children’s knowledge of reconstruction in sen-
with an ACC pronoun could be due to children hearing both
tences with quantificational expressions. English translations of
I+verb (e.g. I do that every day) and me+verb (e.g. Let me do it)
their test sentences are given in (45).
sequences, which could result in children having two compet-
ing constructions for a given verb (e.g. I/me+do) when referring
to themselves.”
Similarly, it has been proposed that a non-adult utterance such
The pair of sentences in (45) are both unacceptable on the read- as He go could result from the omission of the modal can from the
ing where the pronoun he is anaphorically dependent on (bound by) statement He can go, or from the omission of wants to from He wants
the quantificational NP every pirate. So the sentences can’t mean to go (cf. Croker et al., 2003).
that every pirate put a gun in his barrel; it can only mean that As these examples illustrate, the usage-based approach antic-
someone else put a gun in each pirate’s barrel. ipates child language to differ from adult language by being “less
The Guasti and Chierchia study investigated Italian-speaking schematic,” and “simpler with fewer parts.” In general, the usage-
children’s comprehension of PP-preposing Italian sentences like based approach anticipates that children will produce errors of
(46). The semantic representation of (46) is depicted in (47). As omission, rather than errors of commission. Convergence to the
(47) indicates, following reconstruction, the (phonetically empty) adult language is also expected to progress slowly, on the usage-
Subject position has scope over the preposed PP (nel barile di ciascun based approach, from a less articulated linguistic repertoire to one
pirata con cura). As a consequence, anaphoric binding is prohibited that more closely approximates that of adult speakers of the local
in (46), just as in the English example (45a). language. When children begin to form generalizations, these are
based on similarities across constructions.
By contrast, the biolinguistic approach anticipates that chil-
dren’s non-adult productions could be even more filled out than
the productions of adults, including linguistic material that is not
attested in the input. Regardless of the nature of children’s non-
adult productions and interpretations, the biolinguistic approach
anticipates rapid acquisition. First, children’s non-adult responses
The main finding of the study was that 4- and 5-year-old Italian- are highly constrained. They are expected to be compatible with
speaking children rejected the illicit bound pronoun reading 90% of possible human languages, rather than being less articulated vari-
the time. That is, children rejected the sentence as a description of ants of the constructions used by adults. Second, the biolinguistic
a situation in which every pirate put a toy gun in his own barrel. approach anticipates that children will acquire constructions in
In this section we have shown that, contrary to the claim of complexes, amalgamating structures that may look different on
the usage-based account, it is impossible to account for anaphoric the surface. Finally, the biolinguistic approach expects children’s
dependencies solely on the basis of information-theoretic princi- non-adult productions to be overturned in the course of language
ples. Anaphoric dependencies are governed by structural principles acquisition by readily available ‘positive’ evidence.
and, crucially, children demonstrate knowledge of these structural The next two sections report the findings of studies investigating
principles as soon as they can be tested, even in cases in which the acquisition of complex kinds of wh-questions, and sentences
the underlying structure does not match the surface string. Chil- with negation. Both linguistic phenomena have been studied by
dren’s difficulties appear only when they are required to respond advocates of the usage-based approach, and by advocates of the
to sentences that are governed by principles external to syntax. biolinguistic approach. The findings and how the findings are inter-
preted, therefore, are particularly revealing about the alternative
approaches, and permit an assessment of the relative success of
4. Children’s long distance wh-questions both approaches to explain the same observations from child lan-
guage. We begin with the investigation of wh-questions.
Advocates of the usage-based approach contend that children
acquire language by attending to the input and attempting to 4.1. Wh-copying by English-speaking children
formulate constructions that replicate it. Before children have
internalized the constructions of the adult language, they are This section describes an example of a commission ‘error’ that
expected to make certain kinds of errors, but not others. It is is sometimes produced by children acquiring English.8 Although
instructive to consider examples. For example, children sometimes
produce non-adult utterances that lack a finite verb, such as “Her
open it.” According to the usage-based approach, children’s non- 8
We put the term ‘error’ in scare quotes because this term technically misrepre-
adult productions can be explained without recourse to abstract sents the differences between child and adult languages on the generative account.
linguistic principles and can be based on misanalysis of the input We will explain why shortly.
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 133

the sentence structures produced by these children differ from guage draws upon the basic building blocks of Universal Grammar,
those of English-speaking adults, on the biolinguistic approach it including both its principles and its parameters.
is more accurate to refer to children’s productions as non-adult lin- To investigate this, Thornton (1990) used an elicited production
guistic behaviour, rather than as errors. This way of characterizing technique. The technique was used to evoke long-distance ques-
children’s behaviour is in keeping with the Continuity Assumption tions from 21 3- to 5-year-old English-speaking child participants.
(Crain, 1991; Crain and Pietroski, 2001). The present case study is Elicited production tasks have been used successfully to target lin-
instructive in this regard. guistic structures that children rarely produce. In the Thornton
The present example is also instructive because it underscores study, the experimental procedures were designed to encourage
the point that any viable account of language acquisition must children to pose complex information-seeking wh-questions to a
explain entire patterns of children’s linguistic behaviour. It is not puppet. Children posed these question to the puppet, requesting
enough to demonstrate that an error committed by children could the puppet to tell the child, on behalf of the experimenter, what
be produced using the mechanisms invoked by the usage-based it thought about various states of affairs – what it thinks Cookie
account. After all, language acquisition is also characterized by the Monster likes to eat, what it thinks the child had hidden inside a
absence of an unlimited number of non-adult linguistic behaviours. box, who it thinks Grover would like to give a hug, and so on.
Moreover, it turns out that children avoid many potential pitfalls Interestingly, in posing these wh-questions to the puppet,
that would be expected if they were employing the kinds of general- roughly a third of the English-speaking children Thornton inter-
purpose mechanisms that are invoked by the usage-based account. viewed inserted an extra question word into their long-distance
Any viable account of language acquisition is obliged to explain why wh-questions. Two examples of children’s non-adult questions are
certain ‘errors’ never occur. We will show how possible, but not given in (48).
actual errors are avoided because children come forearmed with
the principles and parameters of Universal Grammar.
This section is concerned with the acquisition of wh-questions.
As the term suggests, wh-questions are ones that begin with
words like who and what. There are both simple and complex wh-
questions. A simple wh-question is Who was Elmo talking to? Notice These non-adult wh-questions have a copy of the wh-words who
the ‘gap’ that follows the verb phrase at the end of the question . . . or what in the middle of the questions, so they are also referred
talking to. The question word fills this gap. Nearly all wh-questions to as medial-wh questions. Although medial-wh questions are not
contain an ‘empty’ noun phrase (a ‘gap’) somewhere in the sen- licensed by adult English-speakers, they are an option in other
tence structure.9 The wh-phrase can consist of a ‘bare’ wh-word languages, including dialects of German, colloquial Dutch, Frisian,
(who, what) or a ‘full’ phrase (which boy, what kind of car). In either Afrikaans, and Romani (Du Plessis, 1977; Hiemstra, 1986; Höhle,
case, the wh-phrase is associated with the gap. For this reason, 2000; McDaniel, 1986, 1989). In other words, whether or not long-
wh-phrases are sometimes referred to as ‘fillers,’ and the entire distance questions contain a ‘copy’ of the wh-question word is
question is called a filler-gap dependency. one way in which languages vary. On the biolinguistic account,
An example of a complex wh-question is Who did you say Elmo young English-speaking children who posit medial-wh questions
was talking to? Again, there is a gap following the verb phrase are simply speaking a fragment of a foreign language for a time,
. . . talking to, and the question word who fills the gap. The ques- as sanctioned by the Continuity Assumption. Adopting principles
tion word who and the gap following talking to are a considerable of Universal Grammar, Thornton was able to explain in detail how
distance apart, in different clauses, so these questions are called the foreign-looking questions in (48) emerged in child English, and
long-distance questions. The wh-word in sentence initial position how they were jettisoned in favour of adult questions.
precedes the main clause, which contains a verb such as say or Let us set momentarily aside the issue of the extra copy of the
think that requires an entire sentence as its complement. The com- question word, and focus on the adult versions of these questions:
plement to the verb say in the long-distance wh-question, Who did What do you think Cookie Monster eats? and Who do you think Grover
you say Elmo was talking to? is the embedded sentence, Elmo was wants to hug? The derivation of English wh-questions such as these
talking to. So, the wh-word who in sentence-initial position fills the involves several steps. As we noted, linguistic theory postulates that
gap following the verb phrase talking to in the embedded clause. In question words do not originate in sentence-initial position. Rather,
English, the appearance of a question word in a wh-question usually these question words originate in the object position, following the
corresponds to the presence of a ‘missing’ NP (a gap) somewhere verb in the lower clause. However, question words like who and
else in the sentence. To account for this correspondence, linguis- what do not move to sentence-initial position in one-fell-swoop.
tic theory proposes that most question words are ‘displaced’ from The movement of a question word is more like a local train rather
their original position, leaving behind the gap when they are reposi- than an express train. Before a question word assumes its position
tioned from their extraction site to their landing site in the sentence at the front of a long distance wh-question, it must pass through
structure. an intermediate position, between the upper clause and the lower
clause. As a question word passes through the medial position, it
leaves a copy behind, just as it leaves a ‘gap’ at its site of origin
4.2. Children’s long distance wh-questions
(Fig. 4).
Now we can characterize the differences between languages
A cursory glance at any transcript of infant directed speech leads
that generate medial-wh questions, and ones that do not. The
to the conclusion that long-distance questions are not abundant in
difference is quite superficial; it is whether or not the copy of
young children’s experience. This, in turn, led researchers work-
the intermediate wh-question word is pronounced (see also Rizzi,
ing in the biolinguistic program to carefully study children’s initial
2006). The copy of the question word is not pronounced in adult
attempts at producing long-distance questions, as this promised
English, but it is pronounced in other languages. Children, who
to be extremely revealing about the extent to which child lan-
have little experience producing long-distance questions are thus
presented with two options: to pronounce or not pronounce the
copy. Many children choose the adult English option, and omit the
9
There are two main exceptions to this generalization, wh-questions with why copy. However other children choose the option found in other
and ones with how come. Wh-questions with these wh-phrases do not typically
contain a gap, e.g., Why did John order a pizza?
134 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

is important that in doing their cutting and pasting, children

coordinate not just the linguistic forms involved but also the
conventional communicative functions of these forms, as oth-
erwise they would be speaking creative nonsense. It is also
important that the linguistic structures being cut and pasted
in these acts of linguistic communication are a variegated lot,
including everything from single words to abstract categories
to partially abstract utterance or phrasal schemas.”
At first glance, a cut-and-paste account of children’s non-adult
wh-questions has some intuitive appeal. A closer examination sug-
gests serious problems with this analysis, however. For any account
to be viable, there are two desiderata. First, the account must
Fig. 4. Derivation of long-distance wh-questions.
explain the entire pattern of data that is observed and second, the
account must explain the absence of certain data. In the remainder
languages, and pronounce the copy. These children produce wh- of this section, we show that the usage-based account advanced
questions like those in (48). by Dabrowska et al. (2009) can account for only a limited range
of children’s productions. The account has two main failings. First,
the usage-based account fails to explain the entire pattern of chil-
4.3. A usage-based account of children’s non-adult utterances
dren’s productions. Children produce several medial-wh questions
that the account does not explain. Second, the account predicts
On the usage-based approach, children do not automatically
the appearance of medial-wh questions that are not attested in
have access to the principles of computation that are required to
children’s productions. In other words, the usage-based account
form hierarchical sentence structures. Children have to learn the
overgenerates. This is problematic because children must elimi-
word order of complex structures. It would be surprising to find
nate any non-adult productions that their grammars generate. This
young children producing such complex structures. As Dabrowska
is no easy task given that children lack what is called negative evi-
et al. (2009, p. 1) note:
dence, such as corrective feedback from adults. Previous research
“A number of researchers have claimed that questions and has shown that children are not consistently corrected by adults
other constructions with long distance dependencies (LDDs) are when they make grammatical errors (e.g., Brown and Hanlon, 1970;
acquired relatively early, by age 4 or even earlier, in spite of their Marcus, 1993; Morgan and Travis, 1984). The absence of negative
complexity. . . . Analysis of LDD questions in the input available evidence makes it difficult to explain how children ‘unlearn’ non-
to children suggests that they are extremely stereotypical, rais- adult linguistic structures that their grammars generate. Later, we
ing the possibility that children learn lexically specific templates discuss in detail how the biolinguistic approach contends with the
. . . rather than general rules of the kind postulated in traditional problem. Let it suffice for now to say that, on the biolinguistic
linguistic accounts of this construction.” approach, children’s non-adult productions are highly restricted,
such that they can be expunged from children’s grammars without
These usage-based researchers go on to suggest that children
recourse to negative evidence.
form long-distance wh-questions by drawing on existing ‘tem-
It is worth laying out in more detail what we mean by “overgen-
plates’ for structures they have already learned. The idea is that
eration.” On Dabrowska et al.’s proposal, children should be able to
children simply take two existing templates and put them together.
juxtapose direct questions with “What do you think?” for example.
As Dabrowska et al. (2009, p. 3) state:
Furthermore, since direct questions are much more frequent than
“Interestingly, in their productions of LDDs children sometimes indirect questions children should be more likely to produce ques-
produced questions . . . with a WH word at the beginning of both tions like (49) than ones like (48) or the ones in the quote above.
the main clause and the subordinate clause. But children have not been found to produce questions in which
the auxiliary verb ‘do’ appears in both clauses.
What do you think what is in the box?
What way do you think how he put out the fire?
Thornton and Crain regard such ‘medial WH’ [ones with a copy
between clauses] questions as evidence for the . . . application
of movement . . . Note, however, that such utterances could also In Thornton’s study, however, children as young as three, and
be produced by simply juxtaposing two independent questions one child even younger, had no apparent difficulty producing
(what do you think? + what is in the box?) or an independent a range of long-distance wh-questions, including medial-wh-
question and an indirect question (what way do you think? + questions like (50) and (51).
how he put out the fire?).”
A more general statement of the formation of complex linguistic
structures is provided by Tomasello (2000, p. 77).
“When they have no set expression readily available, they
retrieve linguistic schemas and items that they have previously
mastered (either in their own production or in their comprehen- It is conceivable, as Ambridge and Lieven (2011, p. 306) remark,
sion of other speakers) and then ‘cut and paste’ them together that wh-questions like (50) could have been formed simply by jux-
as necessary for the communicative situation at hand, what I taposing two independent questions (What do you think ? + What’s
have called “usage-based syntactic operations”. Perhaps the first in here?). Example (51) is not amenable to such an account, how-
choice in this creative process is an utterance schema which can ever, because the underlying question fragment... what babies drink
be used to structure the communicative act as a whole, with to grow big? is not an acceptable stand-alone question. The unac-
other items being filled in or added on to this foundation. It ceptability of the wh-question What babies drink to grow big? is
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 135

caused by the absence of the auxiliary verb do. Adding do causes the
question to become well-formed: What do babies drink to grow big?
So, there must be another source of the question fragment... what
babies drink to grow big. Usage-based researchers observed that this
sequence of words is acceptable as the complement of an indirect More often than not, however, children did not insert an extra
question (e.g., Do you know . . . what babies drink to grow big?). This copy of a wh-word in their long-distance wh-questions with full wh-
observation led Dabrowska et al. to propose that children’s medial- phrases. Instead, children produced adult-like long-distance wh-
wh questions like (51) may be created by the juxtaposition of an questions, as in (57).
independent question and an indirect question.
The biolinguistic account of children’s non-adult long-distance
wh-questions first looks beyond English, to other human languages
in order to make sure that children’s non-adult English questions
are UG-compatible. If English-speaking children are simply taking Again, the mechanisms available on the usage-based approach
up an option that is available for other languages, then their out- can easily accommodate illicit wh-questions like (55). These can
put should be similarly constrained. Here, we lay out the case using be created simply by juxtaposition of the question phrase, which
examples from dialects of German. In German, wh-copying is sanc- Smurf do you think and the well-formed English wh-question which
tioned when the question word originates in a finite clause, but not Smurf is holding a toothbrush? The fact that children avoided such
in an infinitive clause, as indicated by the unacceptability of (52). wh-questions with full wh-phrases is further evidence that children
do not access the kinds of mechanisms posited by the usage-based
approach.

5. Negation in child language

Another linguistic phenomenon that has been investigated by

It is also not possible to use a wh-copying structure in sentences both approaches to child language is sentences containing nega-
with full question phrases, such as which man or whose hat. This is tion. Research on sentences with negation began with the seminal
shown in (53). (This example is from Felser, 2004). studies reported in Bellugi (1967) and Klima and Bellugi (1966).
These early studies documented the developmental stages of nega-
tion in three children: Adam, Eve, and Sarah (Brown, 1973). These
were longitudinal studies in which children’s spontaneous produc-
tions were recorded and then transcribed for subsequent analysis.
The analysis found that the three children passed through two
Given the ungrammaticality of (52) and (53) in German, the non-adult stages (Stages 1, 2) before they attained adult-like com-
prediction is that English-speaking children who use wh-copying petence in producing negative sentences (Stage 3).
in many of their questions, will not produce wh-copying struc- At Stage 1, children expressed negation by positioning the nega-
tures for the English counterparts to (52) and (53). This is exactly tive markers ‘no’ or ‘not’ at either end of a word or phrase. Also, the
what happened. The English-speaking children who produced wh- Subject noun phrase was typically omitted. Examples include No sit
questions with a copy of the wh-word in tensed embedded clauses there, Not a teddy bear. Wear mitten no. Stage 1 is often character-
did not pronounce a wh-copy in questions with infinitival embed- ized as a period during which negation is external to the sentence,
ded clauses. Thornton’s (1990) production study elicited questions but this analysis has been challenged (see, e.g., Déprez and Pierce,
with infinitival complements for the verb ‘want’ and found that no 1993; Drozd, 1995; de Villiers and de Villiers, 1985). At Stage 2,
child produced non-adult questions like (54), as indicated by ‘#’. children continue to use no or not, but negation is clearly posi-
tioned sentence-internally. Negation often combines with some
kind of predicate, including main verbs: e.g., He no bite you. I no want
envelope. As these examples illustrate, the main verb is frequently
uninflected at Stage 2, as in Stage 1. Children at this stage begin
Notice that the question in (54) could be generated using the using don’t and can’t (e.g., I can’t catch you, I don’t sit on Cromer cof-
usage-based ‘cut and paste’ mechanism. More specifically (54) fee). However, children’s speech lacks the corresponding positive
could be formed by combining the well-formed adult question, auxiliary verbs can and do. Bellugi (1967) took this to suggest that
What do you want?, and the embedded question structure what to can’t and don’t are unanalyzed (‘fixed’) forms in children’s gram-
do, which is a well-formed substring of the statement I know what mars, rather than being composed from an auxiliary verb and a
to do. negation marker, as in the adult grammar (but see Schütze, 2010).
Another finding favors the biolinguistic account, and is mys- At Stage 2, children’s utterances do not include the auxiliary verb
terious on the usage-based approach. As in German and other does or its negative counterpart doesn’t. It was recently discov-
languages, English-speaking children never produced questions ered that children achieve adult-like mastery of negation, i.e., they
that contained a copy of a full question phrase, such as which Smurf. reach Stage 3, soon after the negative auxiliary verb doesn’t appears
in their speech (Thornton and Tesan, 2007, 2013; Thornton and
Rombough, 2015). As a final observation, children exhibit consid-
erable individual variation in the age at which they reach Stage 3.
Eve reached this stage at 2;02, Adam at 3;02, and Sarah at 3;08.
Instead of wh-questions like (55), the extra ingredient in chil-
dren’s questions with full wh-phrases (e.g., which Smurf) was a
bare wh-word (e.g., who). Some examples with a bare wh-word 5.1. A usage-based account of the acquisition of negation
are presented in (56).
A usage-based account of the acquisition of negation in English
analysed the transcripts of the speech of a child called Brian and
136 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

his mother. The findings and the usage-based analysis are pre- non-adult negative sentences such as ‘not see’ and ‘not run.’ There-
sented in Cameron-Faulkner et al. (2007). The study analysed the fore, a truncation analysis of children’s non-adult utterances falls
extent to which the forms and functions of Brian’s negative sen- victim to contraction.
tences matched those of his mother. Brian’s negative utterances It is clear that input frequency is not the only factor in the
include samples taken from his spontaneous speech between the acquisition of negation, especially at the early stages of language
ages of 2;3–3;4. The Cameron-Faulkner et al. study documents the acquisition. It is also clear that the truncation of adult sentences is
emergence of the negators no, not and the contracted form n’t in not the source of children’s non-adult utterances. Based on these
Brian’s speech, and the correspondence between the emergence of observations, Cameron-Faulkner et al. present the following con-
these forms and their frequency in his mother’s speech. The usage- jecture (p. 273):
based account predicts a strong correspondence between the forms
“it is possible that Brian’s early no V constructions are an amalga-
in the input and the child’s output.
mation of his existing negation strategy (i.e. single word no) with
With only minor variations, the emergence of negative mark-
various entities, states or processes that he wishes to negate. In
ers in Brian’s spontaneous speech conformed to the three stages
this way, Brian’s no V utterances represent a structure-building
charted by Bellugi (1967). Initially Brian’s primary negative marker
approach to multiword negation, as opposed to imitation of
was no. Brian combined no with uninflected verbs in utterances
existing multiword combinations in the input.”
that lacked a Subject noun phrase, e.g., ‘no run’, ‘no move’, ‘no
reach’.10 Although no was also the most frequent negative marker In a review of the same study, Lieven and Tomasello (2008, p.
in the speech of Brian’s mother, she only used no twice in multi- 174) reach a similar conclusion:
word utterances, and it was never followed by an uninflected verb,
“This is an example of creative structure-building at the out-
since this combination is ungrammatical in adult English. There-
set of multiword speech but in complex interaction with the
fore, Brian’s negative utterances did not match the input, as they
frequency of forms in the input and of Brian’s own usage.”
were mainly non-adult linguistic forms. During Stage 1, Brian made
an abrupt shift from using no to signal negation, to using not to As these quotes indicate, input frequency alone cannot account
serve the same functions. As he had done previously with no, Brian for children’s non-adult utterances, even on the usage-based
combined not with uninflected verbs in utterances that also lacked account. It is surprising, however, that Lieven and Tomasello invoke
a Subject noun phrase, yielding utterances such as ‘not see’ and “creative structure-building” to account for children’s non-adult
‘not run’. At Stage 2 (2;9-3;3), Brian continued to use not, but his negative utterances until they develop the ability “to identify more
negative utterances also included the other negative markers can’t specific form-function mappings” (Lieven and Tomasello, 2008).
and don’t. Brian used the negative marker can’t more often than This seems inconsistent with the claim that children’s productions
don’t. However, Brian’s mother used don’t more often than can’t. are “island-like.”
After 3;3, Brian’s negative utterances were reported to be similar In assessing the empirical adequacy of the usage-based
in form and function to those in the input. The conclusion reached approach, it is critical to know what it means for children to be
by Cameron-Faulkner et al. is as follows: “The pattern of negator “creative learners.” Two possible accounts of children’s “creative
emergence was found to follow the frequency of negators in the structure-building” are offered. One is a discontinuity account and
input; that is negators used frequently in the input were the first the second is a continuity account. On the discontinuity account,
to emerge in the child’s speech” (p. 254). The authors’ add a caveat, children form ‘emergent categories’ “that ‘carve up’ conceptual
however: “the findings also indicate creative learning on the part space differently from adults,” according to Cameron-Faulkner et al.
of the child from the earliest stages of multiword negation” (p. Discontinuity is a recurrent theme for the usage-based approach.
254). For example, Tomasello (2000b p. 62) states that “it is obvious to
To explain Brian’s non-adult negative utterances, Cameron- all empirically oriented students of language acquisition that chil-
Faulkner et al. consider the possibility that Brian produced dren operate with different psycholinguistic units than adults, this
truncated versions of sentences produced by his mother. A trun- theoretical freedom to identify these units on the basis of actual
cation analysis has been advanced by usage-based researchers language use, rather than adult-based linguistic theory, is truly
to explain several of children’s non-adult affirmative utterances. liberating.” Although the discontinuity hypothesis may seem obvi-
For example, children at Stage 1 and Stage 2 often produce affir- ous and liberating to advocates of the usage-based approach, it is
mative utterances with uninflected main verbs, such as ‘She eat not the null scientific hypothesis because it introduces unwanted
grapes.’ A truncation account was offered in Tomasello (2000a, degrees of freedom in explaining how children achieve the same
p.240), who suggested that children simply omit the sentence- linguistic competence as adults. An alternative continuity scenario
initial auxiliary verb in adult Yes-No questions, e.g., ‘Does she eat is offered by Cameron-Faulkner et al. On this account, children and
grapes?’ adults have the same “conceptualization of negation.” Cameron-
Cameron-Faulkner et al. consider a similar truncation account of Faulkner et al. suggest that children’s conceptual system “already
children’s non-adult negative sentences like ‘not see’ and ‘not run.’ contains the fine-grained distinctions that underlie the English
They entertain the possibility that these are truncated versions of negation system” (p. 276). The child’s task then is “to discover how
adult negative sentences such as ‘I can not see’ and ‘He did not run’. [the] target language realizes these distinctions.”
They reject this analysis, however, based on the frequency of the In conclusion, the data on children’s production of negation
kinds of negative sentences children encounter in the input. Adult show no clear correspondence between the input to the child
English speakers, including Brian’s mother, rarely combine auxil- and the utterances the child produces. Moreover the usage-based
iary verbs with the negative marker not. Rather, auxiliary verbs are approach fails to provide an account of children’s non-adult nega-
used to host the contracted form of negation, e.g., ‘I can’t see’ and tive utterances, beyond attributing children’s non-adult utterances
‘He didn’t run.’ Such sentences could not be the source of children’s to “creative structure-building” (cf. Lieven and Tomasello, 2008; p.
174). In addition, the usage-based approach has next to nothing
to say about how children make the transition from the non-adult
10
stages of language development to convergence on the adult gram-
Presumably, the negator, no, was used in single word utterances by Brian’s
mother to express PROHIBITION. By contrast, Brian used no followed by an unin-
mar of negation. Let us see how well the biolinguistic approach fares
flected verb to express four functions: FAILURE (No move), PROHIBITION (No touch), in addressing these issues.
REJECTION (No apple), and INABILITY (No reach).
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 137

5.2. A biolinguistic account of the acquisition of negation that children initially construct the most economical syntac-
tic representations available, the default value of the parameter
We now turn to a biolinguistic account of the acquisition of is adverbial negation. As we noted earlier, to incorporate the
negation. All biolinguistic accounts of child language begin with head form of negation, the language learner must construct an
an analysis of the linguistic phenomena in the adult grammar. For additional phrasal projection, beyond that needed for adverbial
English negative sentences, the critical ingredients for deriving neg- negation.
ative sentences in the adult grammar can be highlighted using the Initially, then, children acquiring all human languages are pre-
examples in (58)–(62). dicted to analyse negation as an adverb. If so, then children
acquiring languages with head negation, including English-
speaking children, are predicted to analyse negative markers as
instances of adverbial negation. This will result in non-adult neg-
ative utterances in sentences with head forms of negation (as
illustrated by the difference in acceptability between (59) and
(60)). In particular, treating not like the adverbial never results
in the ungrammatical sentence in (60). However, these non-
adult utterances are expected to be short-lived, because language
learners will encounter abundant positive evidence informing
them that their grammar needs to accommodate a NegP projec-
tion.
In affirmative sentences like (58), the verb carries tense as part Assuming that not, and don’t are (unanalysed) initially as neg-
of the 3rd person ‘s’ morpheme, which also carries number agree- ative adverbs in children’s grammars, then young children are
ment (i.e., the verb form in (58) agrees with a 3rd person singular expected to produce non-adult negative sentences incorporating
Subject NP). The form of the main verb does not change if a negative these negation markers. The specific prediction is that children at
adverb (e.g., never) is added, as (59) illustrates. By contrast, it is not this age should optionally combine these negative markers with
possible to have an inflected main verb like eats (with the 3rd per- the 3rd person ‘s’ morpheme. As we saw in example (50) above,
son ‘s’) if the negative marker not is selected. This is shown by the sentences with the adverb never can be followed by an inflected
ungrammaticality of (60) (marked by ‘*’). Negative sentences with main verb. An example is It never fits. If negation is adverbial in
not can be rescued, however, by inserting the ‘dummy’ auxiliary young children’s grammars, then negative sentences such as It not
verb, do, as in (61). Since the auxiliary verb do carries the 3rd per- fits, and It don’t fits, in addition to their uninflected counterparts It
son ‘s’ morpheme in (61), the main verb remains bare (uninflected). not fit, It don’t fit could, in principle occur. In order to converge on
The main verb also remains bare when the contracted (clitic) form the colloquial adult grammar of English, children need to discover
of negation, n’t, is selected. This is shown in (62). The auxiliary verb that n’t is a head form of negation. Once they have discovered this,
do not only carries the 3rd person ‘s’ morpheme in (62), it also serves they can produce negative sentences with doesn’t, such as It doesn’t
as the host for the clitic form of negation, n’t. So, the negative aux- fit, which is the colloquial form used by adults.
iliary verb doesn’t is decomposed into three parts: do + s + n’t. As we The information that Standard English has a head form of nega-
will see, these features of doesn’t provide critical information for tion in addition to adverbial negative markers is readily available,
children acquiring English, advancing them from Stage 2 to Stage but it requires children to deal with an idiosyncratic aspect of the
3. auxiliary verb system of English, called do-support. To cut a long
Examples (58) to (62) indicate that English has two kinds of story short, the critical evidence for children that English requires
negation. One is adverbial negation, as illustrated in example (59) the construction of a negation phrase (NegP), is the observation that
(Haegeman, 1995; Zanuttini, 2001). Adverbial negation is relatively the clitic form of negation, n’t, is supported by the auxiliary verb do
simple in that it does not require a special, dedicated negation in affirmative statements and in questions (It does fit, Does it fit?)
phrase in the hierarchical structure of English; negative adverbs and in negative sentences (It doesn’t fit). Negative sentences with
are just adverbs like always and usually. By contrast, the second doesn’t are particularly informative because they indicate the tri-
kind of negation in English, illustrated in example (62), requires partite decomposition into do + s + n’t, revealing that the 3rd person
the addition of a special phrasal projection called the Negation ‘s’ morpheme is higher in the syntactic structure than negation. This
Phrase (NegP) (Zeijlstra, 2004). Like other phrasal projections, the led Thornton and Tesan (2007) to propose, and empirically evalu-
NegP has internal structure, which includes a head. Certain neg- ate the prediction that, as soon as a child produces the negative
ative elements, including the clitic form of negation, n’t, usually auxiliary verb doesn’t, that child will have made the transition from
reside in the head position of NegP, so the second form of nega- Stage 2 to Stage 3.
tion in English is called head negation. Example (62) shows that As Thornton and Tesan (2007) note, moreover, once children
the head form of negation, n’t, can be an affix supported by an have acquired doesn’t, all of their non-adult forms of negation are
inflected auxiliary verb such as does (and also by both the cop- predicted to disappear. There have been reports of children pro-
ula verb is and the auxiliary verb is, and by modal verbs, e.g., can, ducing negation followed by an inflected verb, as predicted by
should, will). Thornton and Tesan. For example, Harris and Wexler (1996) con-
Most languages have just one form of negation or the other. ducted a detailed investigation of the transcripts of the spontaneous
Following a survey of 25 languages, Zeijlstra (2004) concluded speech of 10 children (1;06–4;01) using the CHILDES database
that languages can be broadly partitioned into languages in (MacWhinney, 2000). They searched for negative sentences with a
which negation is an adverb and languages in which negation 3rd person subject (such as a name or he or she) and a main verb that
is the head of a phrasal projection (NegP). In view of this lim- required do-support in the adult grammar. The transcripts yielded
ited variation, a parameter for negation, the Negative Concord 54 negative sentences with the negative markers not or no, and 5 of
Parameter was postulated. The Negative Concord Parameter deter- these contained an inflected main verb (see also Croker et al., 2003).
mines where negation is positioned in the sentence structure of The paucity of these negative utterances (less than 10%) led Harris
a language, i.e., whether or not the language requires a NegP and Wexler to consider them to be performance errors. However,
projection. According to Zeijlstra (2004, 2007), learnability consid- the small sample size makes any classification tentative at best.
erations dictate that the parameter has a default setting. Assuming
138 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

In order to increase the sample size of relevant utterances, been found to be reasonably skilled at detecting (local) regularities
Thornton and Tesan (2007, 2013) and Thornton and Rombough in the input. For example, Saffran et al. (1996) found that 8-month-
(2015) conducted two experimental studies using an elicited pro- old infants could exploit statistical regularities in the input to
duction task to target the kinds of sentences that would confirm extract information about ‘word boundaries.’ Infants successfully
or disconfirm their proposal that first, children might initially pro- inferred the existence of boundaries between three-syllable pseu-
duce utterances like It not fits and second, that the negative auxiliary dowords (nonsensical combinations of sound sequences). Those
doesn’t is critical to children’s transition from Stage 2 to Stage 3. One three-syllable sequences that crossed a word boundary were not
was a longitudinal study of four 2-year-old children. The longitu- treated by the child subjects as a ‘word’ during the post-test
dinal study incorporated an elicited production task that targeted phase of the study, because there was a lower probability for such
negative sentences with the third person ‘s’ morpheme. The second sequences to be repeated if they crossed a word boundary than if
study was an elicited production study, using the same technique, they were part of a ‘word.’
with 25 2- and 3-year-old children (mean age = 2;11). In the elic- It is conceivable that children could apply these same skills to
itation component of the longitudinal study, the four 2-year-old extract other kinds of regularities. It has been argued, for example,
child participants produced 497 negative sentences in total with that the input contains relevant features in sufficient abundance to
3rd person subjects combined with a main verb. Ninety-nine of support statistically based acquisition of several seemingly com-
these 497 negative sentences (20%) contained negation followed plex facts about language (MacWhinney, 2004; Pullum and Scholtz,
by an inflected main verb. The majority of these negative sentences 2002). These findings have led some researchers to conclude that
contained the negative marker not (e.g., Minnie Mouse not fits), but children are “perfectly well able to acquire the ‘abstract’ syntactic
some contained don’t (e.g., Minnie Mouse don’t fits). As predicted, concepts that they need to form [structure-dependent] hypotheses
these and many other non-adult negative utterances disappeared through statistical analysis of the speech they hear around them”
from the speech of these children (within 2–3 months) following (Cowie, 2003, p. 192–193).
the appearance of the negative auxiliary doesn’t. On the usage-based approach, the skills children require to form
The larger study by Thornton and Rombough (2015) confirmed generalizations about linguistic input are domain general, rather
the findings of the longitudinal study, and supported the proposal than specific to language. According to Tomasello (2003), children
advanced by Thornton and Tesan that doesn’t triggers children’s utilize the same basic psycholinguistic “perceptual and cognitive
transition to the adult grammar. To investigate this proposal, skills that are employed in other domains as well as language learn-
Thornton and Rombough (2015) divided the 25 child participants ing.” Elaborating on Tomasello’s comment, Cowie (2010) asserts
into two groups, based on whether or not they produced adult-like that children’s analytic skills include “general reasoning skills, such
negative sentences, with the auxiliary doesn’t, in the elicited pro- as the ability to recognize patterns of various sorts in the world, the
duction task. Children (n = 12) who produced at least 5 instances ability to make analogies between patterns that are similar in cer-
of doesn’t were identified as the Advanced group, and the children tain respects, and the ability to perform certain sorts of statistical
(n = 13) who did not produce doesn’t were called the Less Advanced analysis of these patterns.”
group. The children in the Less Advanced group produced a total The biolinguistic approach has a different conception of core
of 4 utterances with doesn’t, and 89 negative utterances with an linguistic phenomena. Linguistic phenomena are not in-and-of
inflected main verb (like Minnie Mouse not/don’t fits). By contrast, themselves core or peripheral; rather core principles underlie ‘nat-
children in the Advanced group produced 228 adult-like negative ural kinds’ of linguistic phenomena. What counts as a natural kind
sentences with doesn’t and only 5 utterances with negation and sometimes includes linguistic phenomena that appear quite dis-
an inflected main verb. Taken together, the findings from both the parate on the surface. An example may be helpful. Consider the
cross-sectional study and from the longitudinal study are com- different expressions that form the class of downward entailing
pelling evidence that the negative auxiliary doesn’t is potentially operators. This includes expressions from several different syn-
a decisive factor in children’s convergence to the adult grammar. tactic categories: prepositions, verbs, adverbs, complex linguistic
structures such as relative clauses with certain quantificational
expressions, and conditional statements. Many downward entail-
6. Scope relations in human languages ing expressions have a negative cast – e.g., the preposition without,
the verb forbid, the adverb never. However, the natural kind also
All human languages contain semi-idiosyncratic constructions includes expressions that do not have a negative cast, such as the
that cannot be derived by universal linguistic principles, and preposition before, the antecedent of conditional statements, if,. . .
that cannot be acquired by the application of innate linguistic then . . ., and the Subject phrase of the universal quantifier every. A
knowledge. On any account of language acquisition, these ‘periph- child who is equipped solely with domain general perceptual abili-
eral’ constructions must be learned. According to the usage-based ties that form generalizations based on analogy or similarity would
account, the same mechanisms that children use to add these be unlikely to uncover several facts about the class of downward
constructions to their language are also used to learn the core phe- entailing expressions. One of the facts is that all English downward
nomena of human languages. The reason is that, on the usage-based entailing expressions license the word any.11 A second fact is that
approach, core linguistic phenomena differ from peripheral phe- disjunction words (English or) generate a ‘conjunctive’ entailment
nomena only in degree – core phenomena are more regular and in downward entailing linguistic contexts.12 As the example illus-
occur more frequently. It follows that core phenomena should be trates, core principles underlie linguistic phenomena that are not
easy to learn. Here is a representative statement by Goldberg (2006, regular or frequent. Children rarely encounter sentences with any
p. 14).
“In fact, by definition the core phenomena are more regular, and
11
tend to occur more frequently within a given language as well. English sentences that license any, bassed on downward entailing expressions
with a negative cast – Bill left without eating any fruit. Bill is forbidden from eating any
Therefore, if anything, they are likely to be easier to learn.” fruit. Bill never eats any fruit. But any is also licensed in the sentences – Bill takes a
pill before eating any fruit. If Bill eats any fruit, he becomes ill.
If core linguistic phenomena were simply constructions that 12
Consider, for example, the sentence Bill never eats apples or oranges. This state-
are more regular and more frequent than more peripheral con- ment with disjunction has two entailments: (1) Bill never eats apples, and (2) Bill
structions, then the usage-based approach would indeed be a never eats oranges. Taken together, they form the ‘conjunctive’ entailment – Bill
contender as an account of language acquisition. Children have never eats apples and Bill never eats oranges.
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 139

or or combined with the universal quantifier, e.g., Every passen- cern that certain sentences engender particular entailments, and
ger who ate any meat. . . , Every passenger who ate chicken or fish. . . others do not.
. However, because downward entailment is a core property of The biolinguistic approach has tried to substantiate the claim
human languages, children are expected to license any in down- that children are able to perform these feats from the earliest
ward entailing linguistic environments and they are expected to stages of language acquisition, before age 3, by pointing to a signif-
generate the entailments that are associated with the disjunction icant body of experimental research that demonstrates children’s
word or in the same environments. This knowledge is expected to adherence to linguistic constraints on form and meaning by age
emerge in child language as soon as children have acquired the 3- or 4-years. Because demonstrating that children have complex
meanings of the relevant words (never, every, any, or and so forth). linguistic knowledge at an early age compresses the timeframe
Core phenomena do not readily lend themselves to statistical anal- for language learning, this reduces the plausibility that children’s
ysis for another reason, namely that children’s knowledge includes linguistic knowledge is acquired using domain general learning
the interpretations that can and cannot be assigned to sentences, mechanisms that enable them to extract regularities in the primary
and is not just based on which words appear together. On the biolin- linguistic data.
guisitic approach, moreover, children are expected to demonstrate In several cases, researchers working within the biolinguistics
such knowledge as soon as they can be tested, by 3- or 4-years-old. approach have documented children’s knowledge of facts for which
In addition, language acquisition is not just a special case of there is (arguably) no decisive evidence in the primary linguistic
induction, or projection beyond one’s experience according to the data. This includes the fact that coreference is not tolerated in sen-
biolinguistic perspective. Consider the following remark from Crain tences like (65). Facts about non-coreference are negative facts, in
and Pietroski (2001, p.161). the sense that children know what sentences like (65) can not mean,
not just what they can mean.
“. . . projecting beyond experience is just one aspect of language
The kind of evidence that would support the acquisition of neg-
acquisition. Children also fail to project beyond their experi-
ative facts is appropriately called negative evidence. The literature
ence in characteristic ways. It is this fact that most impresses
contains several extensive reviews of the availability of negative
nativists. The theoretical problem posed by human language
evidence in children’s experience. These reviews not only discuss
learning is to explain why children project beyond their experi-
the availability of direct negative evidence, such as corrective feed-
ence just so far and no further; the specific ‘angle’ of projection
back when children produce ungrammatical sentences, but they
seems arbitrary (and idiosyncratic to linguistic projection).”
also discuss the availability of various ‘substitutes’ for negative evi-
In the literature, the argument about the specific angle of dence, such as caretakers’ expansions of children’s utterances. It
linguistic projection has primarily focused on language specific does not appear that negative evidence of any kind is available in
constraints on form and interpretation. These constraints include sufficient quantities and at the right time to promote the acquisition
the principles of binding, for example, as discussed in the section on of linguistic knowledge using the kinds of “perceptual and cogni-
anaphoric relations. We saw there that children’s linguistic input tive” mechanisms invoked by the usage-based approach (Morgan
includes sentences like (63) and (64). In both sentences, the pro- and Travis, 1989; Marcus, 1993). In addition to the original Brown
noun he can be interpreted as picking out the same individual as and Hanlon (1970) study, Slobin (1972) reported that children were
the referring expression Papa Smurf, as indicated by the indices. not corrected for ungrammatical utterances in many of the soci-
eties studied by his research group. In a representative review of
the literature, Pinker (1990, p. 217) states the following conclusion.
“When parents are sensitive to the grammaticality of children’s
speech at all, the contingencies between their behaviour and
that of their children are noisy, indiscriminate, and inconsistent
from child to child and age to age.”
If children avail themselves of domain general learning mech-
anisms, then it is conceivable that some children at least would Other researchers have reached the same conclusions (e.g.,
conclude that coreference between the pronoun and the referring Bowerman, 1988; Morgan and Travis, 1989; Marcus, 1993). Even
expression is licensed in sentences like (65), whereas coreference if negative evidence were available, of course, children might not
is ruled out for adults. avail themselves of it. As far as we know, there is no compelling
evidence that children exposed to negative evidence use it to purge
their grammars of incorrect hypotheses (see Newport et al., 1977).

As soon as children can be tested, however, they adhere to the

6.1. A substitute for negative evidence
constraint that prohibits coreference in sentences such as (65) (see
Crain and Thornton (1998) for discussion). This is an example of
Not to be dissuaded, advocates of the usage-based approach
what Crain and Pietroski are referring to when they speak about
have postulated another substitute for (direct) negative evidence.
the specific angle of linguistic projection. To explain the acquisi-
This is the non-occurrence of predicted sentence structures. For
tion of language, we require a theory that enables children to license
example, Cowie (2003, p. 223) asserts that ‘the non-appearance of
coreference in both (63) and (64), but not in (65). On the biolinguis-
a string in the primary data can legitimately be taken as constituting
tic approach, the theory that best accounts for children’s linguistic
negative evidence’. Children could use this substitute for negative
knowledge is one that postulates innate knowledge that is specific
evidence, for example, to expunge the errors that would result from
to language. According to this approach, children come to the task
the application of a structure-independent rule for forming Yes/No
of language acquisition equipped with detailed knowledge of core
questions:
linguistic principles. Not only does the set of core principles explain
why children prohibit coreference in sentences like (65), these prin- “the fact that she has never heard any utterance with the struc-
ciples explain why children as young as 3- or 4-years-old are able ture of *Is that girl who in the jumping castle is Kayley’s daughter
to judge certain sentences to be false, why they are able judge other or *Is that mess that on the floor in there is yours? is evidence that
sentences to be ambiguous, as well as how children are able to dis- strings of that type are not sentences.”
140 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

To exploit this kind of information, however, children must keep near consensus that both children and adults experience difficulty
accurate records of the absence of structures (sentence types) in the in interpreting sentences that require them to compute an ‘inverse’
adult input. Cowie’s example suggests that young children keep a mapping between the surface syntax and the semantic interpreta-
record of the absence of certain kinds of ‘deviant’ relative clauses tion. A direct mapping between sentence word order and semantic
(. . . who in the jumping castle, . . . that on the floor in there) which interpretation is called isomorphism. Therefore, the preference
are themselves embedded inside ‘deviant’ matrix sentences (Is that for a direct mapping between surface word order and semantic
girl . . . is Kayley’s daughter?, Is that mess . . . is yours). Children do interpretation is called the Isomorphism Hypothesis, which can be
not record sentence tokens, however. Rather, children keep records stated as follows.
of sentences types. So, what children would need to notice is the Isomorphism Hypothesis: If a logical expression, A, takes scope
absence of the construction type AUX + RelPro + PP + AUX + NP in over another logical expression, B, in the surface syntax, then A
the input they have encountered, as well as the presence of the also takes scope over B in the semantic interpretation.
construction type AUX + RelPro + AUX + PP + NP. In scope ambiguities involving two logical expressions, a dis-
Researchers working within the biolinguistic approach do not tinction is drawn between the surface scope interpretation and the
attempt to prove that children lack the cognitive skills to keep inverse scope interpretation. On the surface scope interpretation,
records of the presence and absence of such complex structures. the logical expression that comes first takes scope over the one
Rather, they question the plausibility of this substitute for nega- that comes later. On the inverse scope interpretation, the logical
tive evidence as a vehicle used by children in language acquisition. expression that comes later takes scope over one that came ear-
The proposal that children keep accurate records of non-attested lier. According to the Isomorphism Hypothesis, children’s initial
linguistic structure can be challenged, for example, by citing scope assignments are expected to be surface scope interpretations,
conclusions that have been reached in experimental studies of chil- rather then inverse scope interpretations.
dren’s computational resources. The suggestion that children keep Surface scope interpretations are seen to be computationally
detailed records of the complex construction types that they do less complex than inverse scope interpretations. Complexity is
and do not encounter appears to be at odds with studies of human reduced because surface scope interpretations represent 1-to-
memory. We know, for example, that adults can at best recall the 1 mappings between word order and semantic interpretation.
gist of word strings they have just encountered, not the phonologi- On surface scope interpretations, structural units are interpreted
cal or syntactic details of these word strings. Surely children cannot ‘on-line.’ That is, structural units can be semantically composed
be expected to have far superior memories than adults do. More- incrementally, as they are encountered. Sentences that require
over, unless children know in advance which absences to be on inverse scope assignments, by contrast, introduce delays in inter-
the lookout for, they would have to maintain records for all kinds pretation. Assuming that on-line incremental interpretations are
of construction types that can be extracted from the input. These easier for the human sentence processing mechanism (the parser),
records would include much information that will prove irrelevant the usage-based approach is led to formulate concrete predic-
for grammar formation. tions about children’s initial scope assignments, based on the
Isomorphism hypothesis. For example, the Isomorphism Hypothe-
6.2. The isomorphism hypothesis sis predicts that the universal quantifier, every, will take scope over
the negation marker, not, if the universal quantifier precedes the
We turn now to another feature of child language that poses a negation marker in the surface word order. Simply put, the negation
challenge to the usage-based approach. The aspect that is prob- marker will be interpreted in situ.
lematic concerns what are known as scope phenomena. Scope This concrete prediction of the Isomorphism Hypothesis initially
phenomena resist explanation by the kinds of mechanisms the appeared to be confirmed.13 In an early study of negative sentences
usage-based approach attributes to children. However, these phe- with the universal quantifier, Musolino (1998) found that young
nomena can be explained straightforwardly by invoking lexical children rejected sentence (66) if one of the horses did not jump
parameters whose values are ordered in advance by a learning over the fence. This suggests that the universal quantifier was tak-
mechanism known as the Semantic Subset Principle (Crain et al., ing scope over negation, resulting in an interpretation of (66) that
1994; Crain, 2012). The Semantic Subset Principle entreats children can be paraphrased as none of the horses jumped over the fence. This
to initially adopt specific scope assignments, namely ones that can is the surface scope interpretation indicated in (66a), rather than
be adjusted using readily available positive evidence. By encod- the inverse scope interpretation indicated in (66b).
ing these default scope assignments in the parameters of Universal
Grammar, children are prevented from forming erroneous gener-
alizations that they could otherwise make. In this way, children
avoid forming linguistic generalizations that they would need to
retract later. As we have seen, the absence of negative evidence
makes it difficult for children to recover from false starts. On the
Another early finding that is consistent with the Isomorphism
biolinguistic approach, this is not problematic because children’s
Hypothesis was reported in an elicited production study by O’Leary
access to innate linguistic knowledge imposes substantive restric-
and Crain (1994). The study was designed to evoke sentences with
tions on their grammatical hypotheses, preventing the numerous
existential quantifiers from children, such as something and any-
false starts that children could make if their hypotheses were based
thing. In one story, it turned out that one among several dinosaurs
on domain general learning mechanisms. So the acquisition of
could not find anything to eat, but all of the other dinosaurs
scope phenomena illustrates the specific angle of projection taken
managed to find something to eat. The puppet produced a false
by children in the course of language acquisition.
statement about what happened in the story, as illustrated in (67).
On the usage-based approach, children’s assignments of scope
relations between logical expressions, like their assignments of
anaphoric relations, must be based on the surface properties of the
input. The most obvious surface property is word order. There is 13
This conclusion was subsequently challenged by Gualmini (2008). Gualmini
near consensus in the literature that language users and language demonstrated that children access the inverse scope interpretation of scopally
learners prefer a direct linear mapping between surface syntax and ambiguous sentences in pragmatic contexts that satisfy the felicity conditions asso-
semantic interpretation. Putting it the other way around, there is ciated with the use of negation (cf. Musolino and Lidz, 2006).
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 141

In response, children often used the indefinite NP something in the the Isomorphism Hypothesis. The modal paradigm is indicated in
scope of negation. examples (68) and (69).

For adults, the sentences children produced are not accurate

descriptions of the situation. This is because, for adults, the existen-
tial expression something is a Positive Polarity Item. By definition,
Positive Polarity Items (PPIs) take scope over negation at the level When the modal verb può precedes negation (non), as in (68),
of semantic interpretation. For adults, therefore, the sentence the adult speakers of Italian assign the (possible > not) interpreta-
child produced, No, this dinosaur didn’t find something to eat, means tion. When these words appear in the reverse order, as in (69),
that there is something that the salient dinosaur didn’t find to eat adult speakers assign the (not > possible) interpretation. If children
– but that wasn’t what happened in the story. Clearly, children acquiring Italian adopt a domain general strategy based on the dis-
initially differ from adults in their scope assignments. Essentially, tributional analysis– the Isomorphism Hypothesis – they would be
children use some to mean any. Although this finding is consistent expected to easily identify the adult pattern of scope assignment.
with the Isomorphism Hypothesis, it is also consistent with the In Italian, what you see is what you get.
constraint on language learnability we have discussed called the The problem with this acquisition scenario is that child language
Semantic Subset Principle. learners do not know in advance whether they are acquiring Ital-
To avoid learnability problems in the absence of negative evi- ian, versus English or German. Consider the simple negative English
dence, the Semantic Subset Principle dictates that children initially sentence (70). This sentence illustrates that English-speaking
adopt specific values of lexical parameters. In the case at hand, children would be in dire straights if they were to adopt the Isomor-
the Semantic Subset Principle dictates that children initially refrain phism Hypothesis, because the adult interpretation of (70) is the
from interpreting the existential expression someone as a Positive inverse scope interpretation, not the surface scope interpretation.
Polarity Item. Consequently, someone has approximately the same To generate the adult scope assignment for (70), the modal verb can
meaning as anyone in the language spoken by children acquiring must be reconstructed to a position beneath negation, as indicated
English. We will discuss the reason for this momentarily. First, we in (71). The symbol [+R] designates obligatory reconstruction.
describe two linguistic phenomena that undermine the Isomor-
phism Hypothesis.

6.3. A challenge to the isomorphism hypothesis: reconstruction

Despite its simplicity and empirical coverage, children have Based on considerations of language learnability in the absence
been found to respond to several kinds of sentences in ways that of negative evidence, the Semantic Subset Principle dictates that
are not consistent with the Isomorphism Hypothesis. There are two children acquiring all languages initially reconstruct modal verbs
kinds of linguistic phenomena that are not expected on the Iso- that express possibility (Italian può, English can) such that they
morphism Hypothesis. Both phenomena give rise to inverse scope are interpreted within the scope of negation. The reason is that
interpretations. this inverse scope assignment makes sentences true in a narrower
One way inverse scope interpretations can be derived is by range of circumstances than the surface scope interpretation does.
reconstruction. Earlier, we described a syntactic process called Top- If some event may not take place, then this leaves open the possi-
icalization. This process explained the interpretation of ‘topicalized’ bility that the event could take place – but this is ruled out on the
sentences such as John’s mother, he loves dearly. In sentences such ‘not possible’ interpretation. Adopting the Semantic Subset Prin-
as this, the pronoun he and the referring expression John are dis- ciple, Moscati and Crain (2014) predicted that children acquiring
joint in reference. Linguistic theory accounts for this by supposing Italian would initially assign the English-language interpretation
that the topic phrase, John’s mother, reconstructs to Object position (as illustrated in 70) to Italian sentences such as (68). If the inverse
following the verb loves. Following reconstruction, the topic phrase scope interpretation turns out to be children’s initial interpreta-
John’s mother and the pronoun he cannot be assigned coreference; tion of negative sentences like (68), then this would be evidence
they must be disjoint in reference, just as they are in the declarative that children acquiring Italian do not adhere to the Isomorphism
counterpart – He loves John’s mother dearly. Hypothesis. The benefit for children, however, would be that they
Reconstruction has been invoked in the literature to account for would be guaranteed to encounter evidence informing them that
certain scope phenomena. In the resolution of scope ambiguities, the surface scope interpretation is assigned by adults. The evidence
reconstruction ‘lowers’ one scope-bearing expression to a posi- would come in the form of adult utterances of sentences like (68) in
tion beneath another scope-bearing expression in the hierarchical circumstances in which it turned out that Gianni did come after all.
sentence structure, resulting in the inverse scope interpretation. This eventuality would be precluded by children’s inverse scope
If there are inverse scope interpretations that are preferred to sur- assignment. Of course, this circumstance would never eventuate
face scope interpretations, this would undermine the Isomorphism in English, since English-speaking adults assign the inverse scope
Hypothesis. interpretation to sentences like (70).
One case in point is an example from Italian. The example Based on this line of reasoning, Moscati and Crain (2014) pre-
concerns the interpretation that is assigned by Italian-speaking dicted that Italian-speaking children would initially reconstruct the
children and adults to modal expressions in negative sentences. modal verb può to a position beneath negation in interpreting sen-
The interaction between modal verbs and negation is straightfor- tences like (68), despite the fact that adult speakers of Italian do
ward in adult Italian. Scope relations are entirely determined by not. Adults, as we have seen, assign the isomorphic (surface scope)
surface word order. This represents a compelling success story for
142 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

interpretation in sentences where the modal verb può precedes as it does in Mandarin and Japanese. By contrast, conjunction is
negation in the surface syntax. As a consequence, the interpreta- predicted to remain in situ on the Isomorphism Hypothesis.
tion assigned to sentences like (68) by Italian-speaking children
would be completely ungrammatical for adult speakers. This makes
it highly unlikely that children’s interpretation could be based on
the input from adult speakers. This finding would therefore pose a
direct challenge to the usage-based approach, and provide evidence As in the case of Italian modals, the findings do not conform to
against the Isomorphism Hypothesis. the Isomorphism Hypothesis. The relevant findings were obtained
Moscati and Crain (2014) conducted two experimental studies in a study by Notley et al. (2016). This team of researchers inter-
with young Italian-speaking children and adults. These experi- viewed 21 3- to 5-year old English-speaking children (average
ments documented that the scope relations assigned by adult age = 4;9). On a typical trial, a pig had eaten the carrot on offer, but
speakers of Italian were determined by surface word order. not the green pepper. A control group of English-speaking adults
However, Italian-speaking children assigned the inverse scope consistently accepted sentence (74) in this circumstance, whereas
interpretation to negative sentences with epistemic modals, as in the child participants resoundingly rejected (74) in this context
(68). The findings indicated that surface word order did not dic- (98% of the time). Children justified their rejections on the grounds
tate children’s initial assignment of scope relations, whereas adults that the pig had eaten one of the foods. Children’s responses and jus-
do use surface word order to dictate their semantic interpreta- tifications are clear evidence that they assigned the inverse scope
tions. reading to the test sentences. This is another counter-example for
the Isomorphism Hypothesis.

6.4. A second challenge: raising

6.5. Domain specificity
The Isomorphism Hypothesis has been found to make the wrong
prediction in a second linguistic phenomenon. In this case, chil- One of the main differences between the biolinguistic approach
dren have been found to initially ‘raise’ scope-bearing expressions and the usage-based approach concerns domain specificity.
to a position above negation. Again, to avoid subset problems, the According to the usage-based approach, in acquiring a language, it
Semantic Subset Principle compels children to generate inverse suffices to have “perceptual and cognitive skills that are employed
scope interpretations. An example of raising involves children’s in other domains as well as language learning” Cowie (2010).
interpretation of negative statements with conjunction. Consider According to the biolinguistic approach, by contrast, the acquisi-
sentence (72). tion of language is not simply one of many problems of induction
that children solve using general cognitive skills.
There are other features of language that resist explanation if
we invoke mechanisms such as pattern-finding processing or dis-
tributional analysis, which are seen to apply in other cognitive
English-speaking adults accept (72) in three circumstances: (i) domains in addition to language. In this section, we indicate how
when Ted ordered just pasta (ii) when Ted ordered just sushi, scope parameters, in particular, limit the application of learning
and (iii) when Ted did not order either pasta or sushi. When sen- principles in human language.
tence (72) is translated into Mandarin or Japanese, however, adult Consider example (75). For adult English speakers, the existen-
speakers accept the corresponding sentences in only one of these tial expression someone in (75) is a Positive Polarity Item (PPI).
circumstances, namely when Ted did not order either pasta or sushi. The sentence can be paraphrased using the cleft sentence There
Based on these observations, the Semantic Subset Principle dictates is someone that the detectives didn’t find.
that children acquiring all languages will initially raise conjunction
words to take scope over negation. The interpretation that results
resembles that of a cleft sentence in English: ‘It was both sushi
and pasta that Ted did not order.’ This inverse scope reading is
schematically represented in (73). As this paraphrase indicates, someone takes scope over negation
on the interpretation assigned to (75). A graphic depiction of this
interpretation is given in (76). According to linguistic theory, there
are two copies of the existential expression someone in the semantic
representation. The ‘lower’ is pronounced, whereas the ‘upper’ copy
indicates its scope.
In adult Mandarin and Japanese, words for conjunction are Pos-
itive Polarity Items (Crain, 2012). By definition, conjunction words
must be assigned scope over negation at the level of semantic
interpretation, regardless of their position in the surface syntax.
In English and German, by contrast, conjunction words are inter-
preted in situ. In this case, Mandarin and Japanese constitute As noted earlier, children’s productions do not generate the
the subset languages, and English and German are superset lan- same scope assignment as adults do for negative sentences with
guages. the existential expression someone (O’Leary and Crain, 1994). In a
The Semantic Subset Principle predicts that English- and comprehension task, (Musolino et al., 2000), children were found
German-speaking children should initially interpret conjunction as to reject sentence (75) in circumstances in which the detectives
a Positive Polarity Item, as in Mandarin and in Japanese. So children did find someone. Children accepted sentence (75) only if there
acquiring English and German are expected to assign a different wasn’t anyone that the detectives found. That is, children interpret
interpretation than adults do to negated conjunctions like (74). On someone as if it meant anyone, so children’s interpretation of (75)
this parametric account, English- and German-speaking children can be paraphrased by the sentence – The detectives didn’t find any-
are expected to raise conjunction to take scope over negation, just one. The non-adult interpretation on which someone received this
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 143

interpretation was especially prominent in younger children who analyzes surface regularities, like the structure-independent distri-
participated in the study. butional analyzer discussed in Section 2.2. A distributional analyzer
The findings led Musolino et al. to propose that children’s non- ignores the syntactic position of lexical items, just as in Musolino’s
adult analysis of someone represented children’s initial setting of a assertion that “entailment relations between two logical operators
lexical parameter. According to the parameter, someone is a Posi- are not affected by their syntactic position.”
tive Polarity Item (PPI) for adults, but not for children. That is, the The version of the Semantic Subset Principle that Musolino cri-
lexical parameter has two values. The adult value is represented tiques is one advanced in Crain and Thornton (1998, p. 118). It is
as [+PPI]. On this value, the existential expression someone raises important to note, however, that the version of the SSP proposed in
to take scope over negation. On the alternative value, someone is Crain and Thornton (1998) is domain specific, not domain general.
interpreted in situ. This value is represented as [-PPI]. The fact that More specifically, Crain and Thornton state that the SSP is operative
someone is [-PPI] in child language explains why children interpret when “the interpretive component of Universal Grammar makes
someone to have the same meaning as anyone in negative sentences. two interpretations, A and B, available for a sentence, S”. This design
Musolino et al. point out that children’s adoption of the [−PPI] feature prevents children from making false starts in cases of scope
parameter value for someone conforms to the Semantic Subset Prin- ambiguity, where a false start would need to be retracted later, in
ciple (SSP) (Crain et al., 1994; Crain, 2012). A moment’s reflection order for children to converge on the adult language. As we have
indicates that the adult value of the lexical parameter [+PPI] makes seen, recovering from false starts is problematic in the absence of
sentences true in a broader range of circumstances than the [−PPI] negative evidence. To avoid potential learnability problems, the SSP
value. The SSP therefore entreats children to initially assign the guides children’s initial setting of lexical parameters.
[−PPI] value of the lexical parameter. This guarantees that they As the quote from Crain and Thornton (1998) makes clear, both
will encounter positive evidence if they are acquiring languages in values of lexical parameters must be possible in human language.
which adults enforce a polarity restriction on any given existential Therefore, sentence (77) is a viable counter-example to the SSP only
expression. Notice that, on the subset value, someone is interpreted if some possible human language assigns the ‘none’ interpretation
in situ. Because someone appears in Object position, negation takes to such sentences. If there is no language that reconstructs an exis-
wider scope than someone both in the surface order and at the tential expression such as someone from Subject position to a lower
level of semantic interpretation. In other words, children’s inter- position, then there is no potential subset problem, and the SSP is
pretation of sentences like (75) is consistent with the Isomorphism not operative. To state the point differently, the SSP must be consis-
Hypothesis, as well as being consistent with the SSP. tent with the Continuity Assumption. According to the Continuity
Musolino (2006) presents a critique of the Semantic Subset Prin- Assumption, every stage that a child goes through in the course of
ciple (SSP). This critique concludes that the SSP is not the source of language development represents a possible human language (cf.
children’s interpretation of sentences like (75) after all. This leaves Brown, 1973; Crain and Pietroski 2001).
open the possibility that the source of children’s interpretations is We began with the observation that English-speaking chil-
the Isomorphism Hypothesis, so it is important to see if the SSP can dren initially assign a ‘none’ interpretation to sentence (75). We
be rescued from the critique by Musolino. explained this as a consequence of the fact that children assigned
Musolino (2006) argues that the SSP is deficient on both theo- the [−PPI] value to the lexical parameter governing the inter-
retical and on empirical grounds. Space only permits us to discuss pretation of the existential expression someone. On this value of
one empirical challenge (see Moscati and Crain (2014) for a full the parameter, children are expected to interpret the existential
response). One of Musolino’s empirical arguments against the SSP expression someone in situ, regardless of its position in the surface
is based on the interpretation that children and adults assign to syntax. When someone appears in Subject position, sentence (77)
sentences like (77). has the surface scope interpretation, as indicated in (78).

In (77), the existential expression someone occupies the Sub- To falsify the SSP, it must be shown that it is possible for a lan-
ject position. According to Musolino, the SSP entails that someone guage to compel existential expressions to undergo reconstruction,
must reconstruct to a position beneath negation in order to gen- as depicted in (79).
erate the subset ‘none’ interpretation. Following reconstruction,
children would interpret (77) to mean that none of the girls will
ride the merry-go-round. However, the findings from experimental
research show that neither children nor adults assign this interpre-
tation to (77). Instead, someone is interpreted in situ, so someone
takes scope over negation at the level of semantic interpretation, We have seen that reconstruction is not required in English. And,
just as it does in the surface syntax. The empirical findings, there- as far as we know, no language assigns the ‘none’ interpretation to
fore, favour the Isomorphism Hypothesis and are not consistent sentences like (77) (Moscati and Crain, 2014). If not, then there is
with the SSP, according to Musolino. The problem confronting the no lexical parameter from which the SSP selects children’s default
SSP is stated as follows (Musolino, 2006; p.207). setting.
Although the Semantic Subset Principle (SSP) and the Isomor-
“children should initially be restricted to the ‘none’ interpre-
phism Hypothesis often make the same empirical predictions, the
tation of sentences containing . . . existentials and negation,
SSP constrains children’s search space, because it is domain spe-
regardless of the syntactic position of the quantified NPs. This
cific in virtue of being tied to lexical parameters. We have pointed
follows from the fact that entailment relations between two
out two kinds of linguistic phenomena that can be used to assess
logical operators are not affected by their syntactic position.”
the empirical adequacy of these alternative accounts of children’s
This quote indicates that Musolino views the SSP as a general initial scope assignments. Despite the intrinsic appeal of the Iso-
purpose learning principle. On this formulation, the SSP applies in morphism Hypothesis, it is far too general. This is why it is no
all sentences that contain both an existential expression and nega- match for the SSP. The SSP is domain specific; its application is
tion. This formulation of the SSP turns it into a search procedure that limited to lexical parameters. Again, what needs to be explained is
144 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

the specific angle of projection that children take in the course of syntax, but disjunction takes scope over negation at the level of
language acquisition, not just the fact that children project beyond semantic interpretation. This is another example of the inverse
their experience. scope interpretation. Other languages that favour the inverse
scope interpretation of disjunction in negative sentences include
6.6. How languages differ in scope assignments Japanese, Hungarian, Russian, Portuguese, Serbo-Croatian, Slovak,
and Polish.15
According to the Continuity Assumption, child and adult lan-
guages can differ only in ways that adult languages can differ. One
way that adult languages differ is in the assignment of scope rela- 6.7. The Disjunction Parameter
tions to logical expressions. One interpretation of a scope ambiguity
can be strongly favoured in one class of languages, whereas the Universal Grammar is a theory of the initial state of the language
alternative interpretation is strongly favoured in another class of learner. At the initial state, children acquiring all human languages
languages. This kind of cross-linguistic variation holds the poten- are expected to start out with the same default settings of the class
tial to pose a learnability dilemma for children.14 According to the of parameters that hold the potential to pose subset problems. In
biolinguistic approach, however, children come equipped to deal a real sense, all children are therefore expected to speak the same
with the problem, so this aspect of language acquisition is worth language, at least in part. In the case of negative sentences with dis-
discussing in detail. We begin by considering the English sentences junction words, the question arises: Do children start off speaking
(80) and (81). a language in the same class as English or a language in the same
class as Mandarin?
It was predicted by Goro (2004) that all children would ini-
tially speak a language in the same class as English when they
first attempted to interpret negative sentences with disjunction.
The reason is that the different scope assignments for negated dis-
Both of these sentences contain two logical expressions, nega- junctions in sentences like (81) and (82) across languages stand
tion (not) and disjunction (or). Potentially, these logical expressions in a subset/superset relation. On the scope assignment preferred
can be assigned two scope relations. In the case of example (80), by adult English speakers, negative sentences with disjunction are
disjunction takes scope over negation (OR > NOT), so the sentence true in just one circumstance, where both disjunctions are false:
can be paraphrased as Ted didn’t order pasta or Ted didn’t order sushi. NOT A and NOT B. This was illustrated earlier using sentence (81),
The reverse scope assignment is exhibited in example (81). In this Ted did not order sushi or pasta. This sentence is true only if Ted
example, negation takes scope over disjunction (NOT > OR), so the failed to order sushi and failed to order pasta. However, the scope
sentence can be paraphrased as Ted didn’t order pasta and Ted didn’t assignment preferred by adult speakers of Mandarin makes the cor-
order sushi. The scope assignment in the English example (81) con- responding sentence (82) true in a broader range of circumstances.
forms to one of de Morgan’s laws of propositional logic. According Sentence (82) is true for adult speakers of Mandarin when Ted
to this law, a negated disjunction – NOT(A OR B) – entails two failed to order pasta, or when Ted failed to order sushi, or when
negative propositions, NOT(A) and NOT(B). Ted failed to order either pasta or sushi.16 Based on this asymme-
As examples (80) and (81) illustrate, the surface word order try in truth conditions, Goro (2004) reasoned that children would
of English dictates the semantic scope assignments for negative confront a potential learnability dilemma if they initially selected
disjunctions. So English conforms to the Isomorphism Hypothesis, the (superset) scope assignment that is characteristic of Mandarin,
at least in this case. An isomorphism between surface word order OR > NOT.
and scope assignment is not characteristic of other languages, Based on this line of reasoning, Goro predicted that chil-
however. It is not characteristic, for example, of how disjunction dren acquiring all languages would initially assign the (subset)
words are interpreted in negative sentences in Mandarin Chinese. scope relations exhibited in languages like English, NOT > OR. This
Example (82) is the Mandarin Chinese translation of the English assignment of scope relations would mean that children acquiring
example (81). Mandarin would initially judge sentences to be false in certain con-
texts where adult speakers would judge them to be true. Adopting
the Principles and Parameters framework of Universal Grammar,
Goro proposed that the scope assignment of disjunction words was
governed by a lexical parameter, called the Disjunction Parameter.
Adopting different terminology, Goro’s proposal was that disjunc-
Notice that Mandarin and English have the same word order. In tion words were Positive Polarity Items in some languages (e.g.,
example (82), the Mandarin word for negation, méiyŏu, precedes Mandarin) but not in others (e.g., English). As we noted earlier, Posi-
the word for disjunction, huòzhě, just as in the English example tive Polarity Items must take scope over (local) negation at the level
in (81). Nevertheless, adult speakers of Mandarin judge (82) to of semantic interpretation, regardless of the structural relations
express the same meaning as the English cleft sentence in (80), that obtain between disjunction and negation in the surface syntax.
on which disjunction takes scope over negation (OR > NOT), so the
Mandarin sentence (82) means that Ted didn’t order pasta or Ted
didn’t order sushi. In contrast to English, the surface word order 15
It might appear that these languages fail to conform to the relevant law of
in Mandarin does not dictate the semantic interpretation. Nega- propositional logic: ¬(A ∨ B) ⇒ (¬A ∧ ¬B). However, appearances are deceiving.
tion (méiyŏu) takes scope over disjunction (huòzhě) in the surface Because disjunction takes scope over negation in these languages, negation does
not influence the interpretation of disjunction. Disjunction is assigned the same
interpretation in negative sentences as it is in affirmative sentences, and is sub-
ject to the same implicature of ‘exclusivity.’ For evidence that all human languages
14
More technically, subset problems arise when the forms and/or meanings that adhere to certain laws of first order logic, see Crain (2012).
16
are generated on one parameter value asymmetrically entail the forms and/or mean- Consider ambiguous sentence S, with two possible interpretations, A and B. If
ings generated on the other value. Assuming the absence of negative evidence, the interpretation A asymmetrically entails B, then A is true in a subset of the circum-
biolinguistic approach supposes that children initially adopt the parameter value stances that make B true. A is the ‘subset’ interpretation, and B is the ‘superset’
that generates the most restricted set of forms and/or meanings. interpretation.
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 145

Fig. 5. Child and adult patterns of rejection in [+PPI] and in [−PPI] languages.

Setting details aside, we can summarize Goro’s proposal as fol- 6.8. A linguistic universal
lows: disjunction words are associated with a lexical parameter,
such that words for disjunction, OR, are either [+PPI] or [−PPI]. This Although the interpretation of negative sentences with disjunc-
led Goro to predict that children acquiring all human languages tion differs across (adult) languages, there are certain sentence
would initially assign a default value to the lexical parameter, tak- structures that eliminate these cross-linguistic differences. More-
ing disjunction words to be [−PPI]. The default setting of the lexical over, both child and adult speakers of all human languages are
parameter is the ‘subset’ value, so children acquiring languages in expected to assign the same interpretations to these sentences,
which words for disjunction were [+PPI] would encounter adult so these are true linguistic universals. We will go through one
input that would lead them to abandon the default value, in favour example. This example combines several of the concepts we have
of the ‘superset’ interpretation. surveyed in previous sections, but the examples are complex.
As an empirical consequence, adopting the [−PPI] value of the Consider sentence (83). Notice that the disjunction word huozhe
Disjunction Parameter would mean that Mandarin-speaking chil- ‘or’ licenses a Free Choice ‘conjunctive’ inference in (83). So both
dren would interpret the negated disjunction in (82) in the same Mandarin speaking children and adults interpret (83) to mean that
way as English-speaking children and adults interpret the negated Papa Smurf is able to catch bees and Papa Smurf is able to catch
disjunction in (81) Ted didn’t order pasta or sushi. That is, Mandarin- snakes. This is not a logical entailment; it is an inference. This infer-
speaking children were expected to initially take negation to have ence is drawn when disjunction appears in the scope of a modal
scope over disjunction (NOT > OR), just as English-speaking chil- verb like neng ‘can.’
dren and adults do. This prediction runs counter to the usage-based
approach, because the scope assignment that Mandarin-speaking
children are predicted to make is not attested in their input, due
to the fact that adults adopt the [+PPI] value of the Disjunction
Parameter (Fig. 5).
These predictions have been pursued in seven languages so
far: Mandarin, Russian, Japanese, Turkish, German, English, and
Korean. In four of these languages (Mandarin, Russian, Japanese In a recent study by Gao et al. (2016), sentences like (83) were
and Turkish), adult speakers assign the [+PPI] value of the Dis- followed by two kinds of continuations. One continuation was a
junction Parameter. So, adult-speakers of these four languages are full sentence and the other was a fragment of a sentence, where
expected to accept sentences corresponding to the English sentence the disjunction phrase was removed (elided) from the predicate
Ted didn’t order sushi or pasta in contexts in which Ted ordered only phrase. Let us look first at the full sentence continuation, which is
sushi, or only pasta. The critical observation is that children acquir- illustrated in (84).
ing these four languages were predicted to reject these sentences,
in the same fashion as children and adults who are speakers of
languages that adopt the default value of the Disjunction Param-
eter, [−PPI]. As Fig. 5 indicates, this prediction was confirmed.
Children acquiring the four [+PPI] languages consistently rejected
the test sentences, whereas adult speakers of these languages con-
sistently accepted them. Children differ from adults, according to
Goro (2004), because children adhere to the Semantic Subset Princi-
Sentence (84) contains both a negation marker (bu ‘not’) and
ple (Crain et al., 1994; Crain, 2012). The Semantic Subset Principle
disjunction (huozhe ‘or’). We saw earlier that Mandarin-speaking
(SSP) enforces an ordering on the values of parameters in cases
children and adults assign different interpretations to (84). Based
where one value makes a sentence true in a subset of the cir-
on these different interpretations, adults judge (84) to be true,
cumstances that make it true on the other value. The SSP enjoins
whereas children judge it to be false in certain circumstances. One
all children to initially adopt the subset value of the Disjunction
such circumstance is where it is revealed that Sister Smurf cannot
Parameter, regardless of the scope assignment in the local lan-
catch snakes, but can catch bees. Children reject (84) in this context
guage.
because they adopt the default setting of the Disjunction Parameter,
Adult speakers of Mandarin, Russian, Japanese and Turkish typi-
according to which disjunction is [−PPI]. According to this value,
cally interpret disjunction phrases as taking scope over negation. In
disjunction is interpreted in situ, as in English. Therefore, (84) gen-
contrast to adults, children acquiring these languages consistently
erates a conjunctive entailment for Mandarin-speaking children; it
take negation to be the dominant logical operator, taking scope over
entails that Sister Smurf cannot catch bees and that Sister Smurf
disjunction, as shown in Fig. 5.
cannot catch snakes.
146 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

In contrast to children, adult Mandarin-speakers accept (84). For of the phenomena are acquired later than others in the course of
adults, the disjunction word has the [+PPI] value of the Disjunction language development. In fact, acquisitionists working in the gen-
Parameter. According to this value of the parameter, disjunction is erative tradition often make an even stronger hypothesis, namely
forced to take scope over negation at the level of semantic interpre- that children across languages will demonstrate mastery of all of
tation. This yields the ‘not both’ interpretation of sentences such as the relevant phenomena as soon as they can be tested, presumably
(84). So, for adults, (84) is true as long as sister Smurf either can- once they know the meanings of the expressions under investi-
not catch bees or cannot catch snakes. In the context, Sister Smurf gation. There is generally no reason to expect that children need
cannot catch snakes, so sentence (84) is true for adults. months or years to acquire complex linguistic knowledge, in view
Next consider (85). This is the second continuation following of the assistance they receive from Universal Grammar.
sentence (83). In this continuation, the main verb and the disjunc- Early mastery of complex linguistic phenomena compresses the
tion phrase have been elided. The whole verb phrase was used in time frame during which children have access to decisive input
(84): bu neng zhuadao mifeng huozhe xiaoshe ‘not can catch bee or from adult speakers. Therefore, on the biolinguistic approach a use-
snake’. But in (85) only the negation marker and the modal verb ful way of deciding between alternative theories about the course
remain: bu neng ‘not can’. of language acquisition is to investigate the possibility that young
children have knowledge of seemingly complex linguistic phe-
nomena both within the language they are being exposed to, and
across languages. These investigations are especially useful when
the linguistic phenomena are different in character, at least on the
surface. The reason is that the usage-based approach invokes gen-
eral cognitive processes, such as analogy and surface regularities.
The biolinguistic approach, by contrast, anticipates that young chil-
dren will master clusters of disparate-looking phenomena, which
Because the disjunction phrase has been elided, it can no longer are tied together by deep-seated principles of Universal Grammar.
take scope over negation (cf. Crain, 2012). In response to (85), there- It will be instructive to describe the alternatives in more detail.
fore, Mandarin-speaking children and Mandarin-speaking adults According to the usage-based approach, constructions are expected
generate a ‘conjunctive’ entailment (the ‘neither’ interpretation). to be acquired in a piecemeal fashion, especially early in the course
Both child and adult speakers of Mandarin are expected to interpret of language development. The order of acquisition is seen to be
sentences like this in the same way as English-speaking children largely determined by the frequency of the construction in the
and adults. Speakers of both languages are expected to reject (85) input. By eschewing abstract representations as the basis for early
in the context under consideration, where Sister Smurf can catch acquisition, the usage-based approach anticipates that the process
bees, but can not catch snakes. Although sentence (85) contains of amalgamation unfolds only later in the course of acquisition.
disjunction, it licenses a ‘conjunctive’ inference – Sister Smurf can- Moreover, the abstraction processes that underpin the amalgama-
not catch bees and Sister Smurf cannot catch snakes. Gao et al. tion of different constructions when children are 4- or 5- years old
recently interviewed 20 4-year-old Mandarin-speaking children are based on domain general and species general “pattern-finding”
and 20 adults using both full sentence continuations like (84), and cognitive mechanisms. This position is expressed in the following
fragment continuations like (85). As predicted, children and adults quote from Tomasello (2008, pp. 85–86).
produced different responses to the continuation in (84). Children
“Ontogenetically, children hear individual utterances and then
rejected these continuations, whereas adults accepted them. How-
(re-)construct the abstract constructions of a language. All of
ever, both children and adults rejected continuations like (85) over
this is done with general cognitive processes, and universals of
90% of the time in a context in which Sister Smurf was only able to
linguistic structure derive from the fact that people everywhere
catch bees. This is just one example of many in which both cross-
have the same set of general cognitive processes. As noted at the
linguistic and cross-generational differences are negated, leaving
outset, Tomasello (2003) argues that we may segregate these
all language users with the same interpretation.
general cognitive processes into the two overall headings of: (1)
intention-reading, comprising the species unique social cogni-
7. Complete nature as different aspects of one set of tive skills responsible for symbol acquisition and the functional
phenomena dimensions of language, and (2) pattern-finding, the primate
wide cognitive skills involved in the abstraction process.”
From a biolinguistic perspective, the goal of linguistic theory is
the unification or amalgamation of phenomena that look different Cowie (2008/2010) characterizes children’s general reasoning
on the surface, but which are really just different combinations of skills as “. . . the ability to recognize patterns of various sorts in
the same basic building blocks of human languages. Forming gener- the world, the ability to make analogies between patterns that are
alizations that tie together phenomena that appear different on first similar in certain respects, and the ability to perform certain sorts
inspection is the common aim of sciences of all stripes, and has long of statistical analysis of these patterns.”
been at the foundation of linguistic theory. As the physicist Richard Although the usage-based approach credits child language
Feynman put it: . . . the aim is to see complete nature as different learners with powerful reasoning tools, it would not predict that
aspects of one set of phenomena (Feynman 2011, Chapter 2). children, across languages, successfully amalgamate clusters of lin-
Experimental linguistics provides the yardstick for measuring guistic phenomena that are seemingly unrelated on the surface. In
the empirical success of the amalgamations proposed by linguistic this final section, we look at examples of such amalgamation both
theory. Several of the putative deep-seated regularities proposed within and across languages. We present a case study of a cluster of
by linguistic theory have been empirically assessed in studies of apparently unrelated linguistic phenomena that occur within and
child language. These assessments are made using experimen- across languages, which linguistic theory has attempted to account
tal techniques designed to unveil young children’s knowledge of for using just a few basic concepts and inferential mechanisms.
the relevant phenomena. A proposal about the amalgamation of The description of the theory is followed by a review of experi-
disparate-looking phenomena is confirmed if the phenomena are mental studies of these phenomena in children acquiring Mandarin
acquired as a package by young language learners. The alterna- Chinese. We have chosen to use Mandarin to showcase children’s
tive (i.e., disconfirmation) would be the finding that one or another unification of disparate looking linguistic properties as an exam-
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 147

ple for several reasons. First, Mandarin is historically unrelated to these lexical items are built from the same basic building blocks,
English. If it can be shown that children acquiring Mandarin draw despite their differences for adult speakers.
upon the same linguistic toolkit as children acquiring English, this
would be compelling evidence that the biolinguistic approach is on 7.2. Free choice inferences
the right track. Mandarin is also useful because of its special linguis-
tic properties. We will demonstrate that the language particular The final topic is another kind of amalgamation that takes place
properties of Mandarin are merely different ways of assembling both within and across languages. It turns out, not accidently,
the same basic linguistic structures, as compared to English, or any that English any, its Mandarin counterpart renhe and Mandarin
other language. question words such as shenme ‘what’ are all licensed by the
negative quantificational phrase, nobody/meiyouren: Meiyouren chi
7.1. Disjunction as an existential expression renhe/(shenme) shuiguo ‘Nobody ate any fruit.’ In this linguis-
tic environment, these expressions are labeled Negative Polarity
We begin by pointing to the theoretical overlap between state-
Items (NPIs). In the same linguistic environment, the Mandarin
ments with disjunction, and the corresponding statements with
disjunction word huozhe and its English counterpart or generate
existential expressions. In human languages, as in logic, disjunction
a conjunctive interpretation. For example, the English sentence
words (English or, Mandarin huozhe) and existential expressions
Nobody ate an apple or an orange entails that nobody ate an apple
(English any, Mandarin renhe) are intimately linked. Suppose Ted
and it entails that nobody ate an orange. When English any and
is choosing from a limited menu, with only two main offerings,
Mandarin renhe appear in other structures, however, they license
pasta and sushi. If Ted decides against pasta, but orders sushi,
free choice inferences.
then English translations of sentences (86) and (87) will both be
Free choice inferences are licensed in English when any appears
judged to be false. Moreover, the Mandarin sentences are both
in sentences with a modal verb, including the epistemic modal can
false for Mandarin-speaking children. In contrast to children, how-
(meaning is able to) or the deontic modal may (meaning is allowed
ever, Mandarin-speaking adults judge (86) to be true because Ted
to). This is illustrated in example (88). Again, in a finite domain, with
ordered sushi, but not pasta.
just one green car and one red car, example (88) is logically equiv-
alent to the disjunctive statement (89). Both of these sentences can
be paraphrased using conjunction, so they can both be paraphrased
as follows: Kung Fu Panda can/may push the green car, and Kung Fu
Panda can/may push the red car; he is free to choose which car to push.
One formal algorithm for computing free choice inferences is called
recursive exhaustification.17

Sentence (86) is false for children acquiring any language, as

far as we know. The reason is that is the default value of the Dis-
junction Parameter is [−PPI], and this value results in a conjunctive
entailment (the ‘neither’ interpretation). This is also the value of the The observation that disjunction words license free choice infer-
Disjunction Parameter for adult English speakers. The Mandarin ences is surprising.
and English sentences in (86) are logically equivalent to the sen-
tences in (87) as long as disjunction is interpreted within the scope
of negation. On this scope assignment, both (86) and (87) are called
∃-items. They receive this designation because both huozhe/or and
As (90) illustrates, disjunction phrases do not typically license
any/renhe are variants of the existential quantifier, ∃. If disjunction
free choice inferences. These inferences are licensed only when a
is [+PPI], however, then disjunction takes scope over negation, as
disjunction phrase is combined with certain linguistic expressions,
in adult Mandarin.
such as the modal verbs can and may. In fact, adult English speakers
Across languages, children’s interpretation of sentences like (86)
have the reverse intuition about (90). For most adults, (90) means
and (87) reveals their knowledge of the unity between disjunction
that Kung Fu Panda did NOT push both cars. This ‘exclusivity’ (‘not
and existential expressions such as Mandarin renhe and English
both’) inference is effected by the fact that (90) contains or rather
any. The challenge for the usage-based approach to language acqui-
than and. If Kung Fu Panda had pushed both cars, then and would
sition is to explain how Mandarin-speaking children could have
be the operative logical connective in (90), since the use of and
discovered the unity of disjunction and existential expressions
would have conveyed the facts more directly. The use of or, there-
based on their linguistic experience, given that adult speakers of
fore, invites readers to infer that Kung Fu Panda did not push both
Mandarin judge (86) to be true and (87) to be false, whereas chil-
of the cars.
dren judge both sentences to be false. It is unlikely, therefore, that
children could have discovered that disjunction and existential
expressions are both ∃-items using domain general cognitive mech-
anisms based on similarities in the distributions of lexical items. 17
Recursive exhaustification involves two applications of a ‘exhaustfication’ oper-
Because Mandarin-speaking adults assign the [+PPI] feature to dis- ator, ONLY. First, ONLY factors in subdomain alternatives and their associated
inferences. Take statement (89), which we render as ♦[G ∨ O], where G stands for
junction, sentences (86) and (87) do not pattern in the same way for
green, and O for orange. The subdomain alternatives include {♦G, ♦O}. For alterna-
them. But this means that the parental input obscures the underly- tive ♦G, the first exhaustification ONLY(♦G) generates the inference [♦G ∧ ¬♦O]. For
ing generalization – that disjunction and existential expressions are alternative ♦O, it generates [♦O ∧ ¬♦G]. The second application of ONLY disposes
cut from the same cloth. Therefore, the finding that children inter- of alternatives (with their associated inferences) that are informationally stronger
pret sentences like (86) and (87) as equivalent in meaning must than the original assertion ♦[G ∨ O]. Both of the propositions generated at the first
step are stronger than the assertion, so they are inferred to be false. Therefore, the
be explained without recourse to children’s pattern-finding abili- second exhaustification yields ♦[G ∨ O] ∧ ¬[♦G ∧ ¬♦O] ∧ ¬[♦O ∧ ¬♦G] or, equiva-
ties. These findings are consistent with the biolinguistic approach, lently, ♦[G ∨ O] ∧ [♦G ↔ ♦O]. Taken together, the original assertion ♦[G ∨ O] and the
because this approach anticipates that children will postulate that inference [♦G ↔ ♦O] entitle us to conclude ♦G ∧ ♦O.
148 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149

Free choice inferences are not just a property of English. Man- References
darin licenses free choice inferences in sentences with renhe ‘any’ Ambridge, B., Lieven, E.V., 2011. Child Language Acquisition: Contrasting
and in sentences with disjunction huozhe ‘or’, just as English does. Theoretical Approaches. Cambridge University Press, Cambridge.
Moreover, Mandarin licenses free choice inference with ques- Ambridge, B., Rowland, C., Pine, J., 2008. Is structure dependence an innate
constraint: experimental evidence from children’s complex-question
tion words such as shenme ‘what.’ We illustrate this in examples
production. Cogn. Sci. 32, 222–255.
(91)–(93) using one of the most intriguing expressions in Mandarin, Ambridge, B., Pine, J.M., Lieven, E.V., 2014. Child language acquisition: why
the adverbial quantifier dou ‘all/always.’18 universal grammar doesn’t help. Language 90 (3), e53–e90.
Bellugi, U., 1967. The Acquisition of the System of Negation in Children’s Speech.
Harvard University, Boston, MA, Unpublished doctoral dissertation.
Berwick, R.C., Pietroski, P., Yankama, B., Chomsky, N., 2011. Poverty of the stimulus
revisited. Cogn. Sci. 35 (7), 1207–1242.
Berwick, R.C., Chomsky, N., Piattelli-Palmarini, M., 2012. Poverty of the stimulus
stands: why recent challenges fail. In: Piattelli- Palmarini, M., Berwick, R.C.
(Eds.), Rich Languages from Poor Inputs. Oxford University Press, Oxford, UK,
pp. 19–42.
Berwick, R.C., 1985. The Acquisition of Syntactic Knowledge. MIT Press, Cambridge,
MA.
Bowerman, M., 1988. The ‘no negative evidence’ problem: how do children avoid
constructing an overly general grammar? In: Hawkins, J. (Ed.), Explaining
Language Universals. Oxford, Basil, Blackwell, pp. 73–101.
Boyd, J.K., Goldberg, A.E., 2011. Learning what not to say: the role of statistical
preemption and categorization in a-adjective production. Language 87 (1),
55–83.
As expected, the Mandarin expressions renhe ‘any’ and shenme Brown, R., Hanlon, C., 1970. Derivational complexity and order of acquisition in
‘what’ in sentences (92) and (93) have the same meaning. This child speech. In: Hayes, J.R. (Ed.), Cognition and the Development of Language.
underscores the conclusion that these expressions have a common Wiley, New York, pp. 11–54.
Brown, R., 1973. A First Language: The Early Stages. Allan and Unwin, London.
source. Despite the apparent differences between these expres- Cameron-Faulkner, T., Lieven, E.V., Theakston, A., 2007. What part of no do
sions in ordinary sentences, looking at more complex examples children not understand? A usage-based account of multiword negation. J.
brings to light evidence in favour of the unified account, accord- Child Lang. 34, 251–282.
Chien, Y.-C., Wexler, K., 1990. Children’s knowledge of locality conditions in
ing to which these expressions are different instantiations of the binding as evidence for the modularity of syntax and pragmatics. Lang. Acquis.
existential quantifier, ∃, in both English and in Mandarin. 1, 225–295.
Chomksy, N., 1965. Aspects of the Theory of Syntax. MIT Press, Cambridge, MA.
Chomsky, N., 1980. Rules and Representations. Columbia University Press, New
8. Conclusion York.
Chomsky, N., 1981. Lectures on Government and Binding. Foris, Dordrecht.
Chomsky, N., 1988. Language and Problems of Knowledge: The Managua Lectures,
This chapter has compared the usage-based approach to lan- vol. 16. MIT Press, Cambridge, MA.
guage acquisition with the biolinguistic approach. We reported the Chomsky, N., 2007. Of minds and language. Biolinguistics 1, 009–027.
Chomsky, N., 2012. Poverty of the stimulus: willingness to be puzzled. In:
findings of experimental studies in several areas of child language, Piattelli-Palmarini, M., Berwick, R.C. (Eds.), Rich Languages from Poor Inputs. ,
in most cases where both approaches have conducted research. p. 61.
It seems clear to us that the biolinguistic account is both more Chomsky, N., 2013. The Dewey lectures 2013: what kind of creatures are we? J.
Philos. 12, 645–700, Volume CX.
descriptively adequate, and more explanatory. We offered several Conroy, A., Takahashi, E., Lidz, J., Phillips, C., 2009. Equal treatment for all
examples of linguistic phenomena where children have reached antecedents: how children succeed with Principle B. Ling. Inq. 45, 446–486.
conclusions that are not a direct reflection of the input, includ- Cowie, F., 2003. What’s Within?: Nativism Reconsidered. Oxford University Press,
New York.
ing children’s productions and their understanding of language.
Cowie, F., 2010. Innateness and language. In: Edward Zalta, N. (Ed.), The Stanford
We have also discussed several examples of children’s non-adult Encyclopedia of Philosophy. , summer 2010 edition, URL: http://plato.stanford.
productions and comprehension that resist explanation by the gen- edu/archives/sum2010/entries/innateness-language/>.
Cowie, F., 2008/2010. Innateness and language. In: Stanford Encyclopedia of
eral cognitive processes adopted by the usage-based approach, but
Philosophy.
ones that are expected on the biolinguistic approach. The biolin- Crain, S., Nakayama, M., 1987. Structure dependence in grammar formation.
guistic approach expects there to be differences between child and Language 63, 522–543.
adult language. However, when child and adult language differ, Crain, S., Pietroski, P., 2001. Nature, nurture and universal grammar. Ling. Philos.
24, 139–186.
children are expected to invoke structures from a possible human Crain, S., Thornton, R., 1998. Investigations in Universal Grammar: A Guide to
language, just not the one the child is exposed to. Finally, children Experiments on the Acquisition of Syntax and Semantics. MIT Press,
were found to amalgamate linguistic phenomena that look differ- Cambridge, MA.
Crain, S., Ni, W., Conway, L., 1994. Learning, parsing and modularity. In: Clifton, C.,
ent on the surface, rather than forming generalizations based on Frazier, L., Rayner, K. (Eds.), Perspectives on Sentence Processing. Lawrence
similarity, analogy, or distributional analysis. Finding unity when Erlbaum Associates, Inc., Hillsdale, New Jersey.
confronted by diversity is the hallmark of the biolinguistic approach Crain, S., 1991. Language acquisition in the absence of experience. Behav. Brain Sci.
14, 597–650, http://dx.doi.org/10.1017/S0140525X00071491.
to language acquisition. Crain, S., 2012. The Emergence of Meaning. Cambridge University Press, Cambridge.
Croker, S., Pine, J.M., Gobet, F., 2003. Modelling children’s negation errors using
probabilistic learning in MOSAIC. In: Detje, F., Dörner, D., Schaub, H. (Eds.),
Acknowledgement Proceedings of the Fifth International Conference on Cognitive Modeling.
Bamberg: Universitäts-Verlag, pp. 69–74.
This research was funded by the Australian Research Council Cen- Déprez, V., Pierce, A., 1993. Negation and functional projections in early grammar.
Ling. Inq. 24, 25–67.
tre of Excellence in Cognition and its Disorders (CE110001021) www. Dabrowska, E., Rowland, C., Theakston, A., 2009. The acquisition of questions with
ccd.edu.au long-distance dependencies. Cogn. Ling. 20 (3), 571–598.
Deutsch, W., Koster, C., Koster, J., 1986. What can we learn from children’s errors in
understanding anaphora? Linguistics 24, 203–225.
de Villiers, J., de Villiers, P., 1985. The acquisition of English. In: Slobin, D. (Ed.), The
18
English and Mandarin differ in how quantification is realized. In English, quan- Crosslinguistic Study of Language Acquisition, Vol 1: The Data. Lawrence
tificational expressions can appear in two syntactic positions, either as Determiners Erlbaum, Hillsdale, NJ, pp. 27–140.
or as Adverbs. The universal quantifier every is a Determiner, and so are the exis- Drozd, K., 1995. Pre-sentential negation as metalinguistic exclamatory sentence
tential indefinites some and a. The expressions always, often, seldom, necessarily, and negation. J. Child Lang. 22, 583–610.
Du Plessis, H., 1977. Wh movement in Afrikaans. Ling. Inq. 8, 723–726.
sometimes are Adverbs of quantification. In Mandarin, quantificational expressions
Elbourne, P., 2005. On the acquisition of principle B. Ling. Inq. 36, 333–365.
are always Adverbs, since Mandarin lacks Determiners altogether.
 S. Crain et al. / Neuroscience and Biobehavioral Reviews 81 (2017) 120–149 149

Everaert, M.B., Huybregts, M.A., Chomsky, N., Berwick, R.C., Bolhuis, J.J., 2015. Musolino, J., 2006. On the semantics of the subset principle. Lang. Learn. Dev. 2 (3),
Structures, not strings: linguistics as part of the cognitive sciences. Trends 195–218.
Cogn. Sci. 19 (12), 729–743. Newport, E.L., Gleitman, H., Gleitman, L.R., 1977. Mother I’d rather do it myself: the
Felser, C., 2004. Wh-copying, phases, and successive cyclicity. Lingua 114, 543–574. contribution of selected child listener variables. In: Snow, C.E., Ferguson, C.A.
Feynman, R.P., 2011. Six Easy Pieces. Basic Books. (Eds.), Talking to Children. Cambridge University Press, Cambridge.
Fodor, J., 1983. Modularity of Mind. MIT Press, Cambridge, MA. Notley, A., Zhou, P., Crain, S., 2016. Children’s interpretation of conjunction in the
Gao, N., Crain, S., Zhou, P., Thornton, R., 2016. The interpretation of disjunction scope of negation in English and Mandarin: new evidence for the Semantic
involving negative linguistic contexts in Mandarin Chinese. In: Paper Subset Maxim. Appl. Psycholinguist. 37, 867–900.
Presented at Disjunction Days: Theoretical and Experimental Perspectives on O’Leary, C., Crain, S., 1994. Negative polarity (a positive result) and positive
the Semantics And Pragmatics of Disjunction, Zentrum für Allgemeine polarity (a negative result). In: Paper Presented at the 18th Annual Boston
Sprachwissenschaft (ZAS Berlin), June 2–3. University Conference on Language Development, Boston, MA.
Gavruseva, E., Thornton, R., 2001. Getting it right: acquisition of whose-questions Philip, W., Coopmans, P., 1996. The role of lexical feature acquisition in the
in child English. Lang. Acquis. 9, 229–267, http://dx.doi.org/10.1207/ development of pronominal anaphora. In: Philip, W., Wijnen, F. (Eds.),
S15327817LA0903 2. Amsterdam Series on Child Language Development, 5. Instituut Algemene
Goldberg, A., 2003. Constructions: a new theoretical approach to language. Trends Taalwetenschap 68, Amsterdam, The Netherlands.
Cogn. Sci. 7, 219–224. Pinker, S., 1984. Language Learnability and Language Development. Harvard
Goldberg, A., 2006. Constructions at Work: The Nature of Generalizations in University Press, Cambridge, MA.
Language. Oxford University Press, New York. Pinker, S., 1990. Language acquisition. In: Osherson, D.N., Lasnik, H. (Eds.), An
Goro, T., 2004. The Emergence of Universal Grammar in the Emergence of Invitation to Cognitive Science, vol. 1. MIT Press, Cambridge, MA, pp. 199–241.
Language: the Acquisition of Japanese Logical Connectives and Positive Pinker, S., 1994. The Language Instinct: How the Mind Creates Language. Harper
Polarity. Unpublished Doctoral Dissertation. University of Maryland, College Perennial Modern Classics, New York, NY.
Park, MD. Pullum, G., Scholtz, B., 2002. Empirical assessments of stimulus poverty
Grodzinsky, Y., Reinhart, T., 1993. The innateness of binding and coreference. Ling. arguments. Ling. Rev. 19, 9–50.
Inq. 24, 69–101. Reinhart, T., Reuland, E., 1993. Reflexivity. Ling. Inq. 24, 657–720.
Gualmini, A., Crain, S., 2005. The structure of children’s linguistic knowledge. Ling. Reuland, E., 2001. Primitives of binding. Linguistic Inquiry 32 (3), 439–492.
Inq. 36, 463–474. Reuland, E., 2011. Anaphora and Language Design. MIT Press, Cambridge, MA.
Gualmini, A., 2008. The rise and fall of Isomorphism. Lingua 118, 1158–1176. Rizzi, L., 2006. Grammatically-based target-inconsistencies in child language. In:
Guasti, M.T., Chierchia, G., 2000. Backward versus forward anaphora: Deen, K.U., Nomura, J., Schulz, B., Schwartz, B.D. (Eds.), Proceedings of the
reconstruction in child grammar. Lang. Acquis. 8, 129–170. Inaugural Conference on Generative Approaches to Language
Höhle, T., 2000. The w.w construction: appositive or scope indicating? In: Lutz, U., Acquisition–North America, Honolulu, HI. Storrs, MA: University of
Müller, G., von Stechow, A. (Eds.), Wh-scope Marking. John Benjamins, Connecticut Occasional Papers in Linguistics, 4, pp. 19–49.
Amsterdam, pp. 249–270. Saffran, J., Aslin, R., Newport, E., 1996. Statistical learning by 8-month-old infants.
Haegeman, L., 1995. The Syntax of Negation. Cambridge University Press, Science 274, 1926–1928.
Cambridge. Saxton, M., 2010. Child Language: Acquisition and Development. Sage Publications,
Harris, T., Wexler, K., 1996. The optional-infinitive stage in child language: London.
evidence from negation. In: Clahsen, H. (Ed.), Generative Perspectives on Schütze, C., 2010. The status of nonagreeing don’t and theories of optional
Language Acquisition. John Benjamins, Amsterdam, pp. 1–42. infinitives. Lang. Acquis. 17, 235–271, http://dx.doi.org/10.1080/10489223.
Heycock, C., Kroch, A., 2002. Topic, focus, and syntactic representations. Cascadilla 2010.509270.
Press: Somerville, MA In: Proceedings of WCCFL, 21, pp. 141–165. Slobin, D., 1972. Children and language: they learn the same way all around the
Hiemstra, I., 1986. Some aspects of wh-questions in Frisian. North-Western world. Psychol. Today 6, 71–82.
European Language Evolution (NOWELE), 8, 97–110. Sportiche, D., 2006. Reconstruction, binding and scope. In: Everaert, M., van
Hyams, N., 1986. Language Acquisition and the Theory of Parameters. Kluwer, Riemsdijk, H. (Eds.), The Blackwell Companion to Syntax, vol. I–V. Blackwell,
Dordrecht. Oxford.
Jakubowicz, C., 1984. On markedness and binding principles. In: Proceedings of the Thornton, R., Rombough, K., 2015. The syntax-PF interface in children’s negative
14th Annual Meeting of the Northeast Linguistics Society (NELS), Amherst, sentences. Lang. Acquis. 22, 132–157, http://dx.doi.org/10.1080/10489223.
MA: GLSA, pp. 154–182. 2014.943901.
Karmiloff-Smith, A., 1992. Beyond Modularity: A Developmental Perspective on Thornton, R., Tesan, G., 2007. Categorical acquisition: parameter setting in
Cognitive Science. MIT Press/Bradford Books, Cambridge, MA (Reprinted 1995). Universal Grammar. Biolinguistics 1, 49–98.
Kirjavainen, M., Theakston, A., 2009. Can input explain children’s me-for-I errors? Thornton, R., Tesan, G., 2013. Sentential negation in early child English. J. Ling. 49
J. Child Lang. 36, 1091–1114. (2), 367–411.
Klima, E., Bellugi, U., 1966. Syntactic regularities in the speech of children. In: Thornton, R., Wexler, K., 1999. Principle B, VP Ellipsis and Interpretation in Child
Lyons, J., Wales, R. (Eds.), Psycholinguistic Papers. Edinburgh University Press, Grammar. MIT Press, Cambridge, MA.
Edinburgh, pp. 183–208. Thornton, R., Kiguchi, H., D’Onofrio, E., 2015. Clefts and reconstruction in
Koster, C., 1993. Errors in Anaphora Acquisition. Led, Utrecht: Onderzoeksinstituut English-speaking children’s grammars. In: Paper Presented at the 40th Boston
voor Taal en Spraak Rijksuniversiteit Utrecht, Utrecht. University Conference on Child Language Development, November 13–15.
Lieven, E.V.M., Tomasello, M., 2008. Children’s first language acquisition from a Thornton, R., 1990. Adventures in Long-distance Moving: the Acquisition of
usage-based perspective. In: Robinson, P., Ellis, N.C. (Eds.), Handbook of Complex Wh-questions. University of Connecticut, Storrs, Unpublished
Cognitive Linguistics and Second Language Acquisition. Rutledge, New York, doctoral dissertation.
pp. 168–196. Tomasello, M., 2000a. Do young children have adult syntactic competence?
Lieven, E.V.M., 2010. Input and first language acquisition: evaluating the role of Cognition 74, 209–253.
frequency. Lingua 120, 2546–2556. Tomasello, M., 2000b. First steps toward a usage based account of language
MacWhinney, B., 2000. The CHILDES Project: Tools for Analyzing Talk. Lawrence acquisition. Cogn. Ling. 11, 61–82.
Erlbaum Associates, Mahwah, NJ. Tomasello, M., 2003. Constructing a Language: A Usage-based Theory of Language
MacWhinney, B., 2004. A multiple process solution to the logical problem of Acquisition. Harvard University Press, Cambridge, MA.
language acquisition. J. Child Lang. 31, 883–914. Tomasello, M., 2008. How Are Humans Unique?, 25. Times Magazine, New York
Marcus, G.F., 1993. Negative evidence in language acquisition. Cognition 46, 53–85. (May).
Matthews, D., Lieven, E., Theakston, A., Tomasello, M., 2009. Pronoun Valian, V., Solt, S., Stewart, J., 2009. Abstract categories or limited-scope formulae?
co-referencing errors: challenges for generativist and usage-based accounts. The case of children’s determiners. J. Child Lang. 36 (04), 743–7778.
Cogn. Ling. 20, 599–626. Yang, C.D., 2002. Knowledge and Learning in Natural Language. Oxford University
McDaniel, D., 1986. Conditions on Wh-chains. City University of New York, New Press on Demand.
York, Unpublished doctoral dissertation. Yang, C., 2013. Ontogeny and phylogeny of language. Proc. Natl. Acad. Sci. 110 (16),
McDaniel, D., 1989. Partial and multiple wh-movement. Nat. Lang. Linguistic 6324–6327.
Theory 7 (4), 565–604. Yang, C., 2015. For and against frequencies. J. Child Lang. 42, 287–293, http://dx.
Morgan, J., Travis, L., 1989. Limits on negative information in language input. J. doi.org/10.1017/S0305000914000683.
Child Lang. 16, 531–552. Zanuttini, Rafaella, 2001. Sentential negation. In: Mark, Baltin, Chris, Collins (Eds.),
Moscati, V., Crain, S., 2014. When negation and epistemic modality combine: the The Handbook of Contemporary Syntactic Theory. Blackwell, Malden, MA, pp.
role of information strength in child language. Lang. Learn. Dev. 10 (36), 511–535.
345–380, http://dx.doi.org/10.1080/15475441.2014.880640. Zeijlstra, H.H., 2004. Sentential Negation and Negative Concord. University of
Musolino, J., Lidz, J., 2006. Why children aren’t universally successful with Amsterdam, Amsterdam, The Netherlands, Unpublished doctoral dissertation.
quantification. Linguistics 44 (4), 817–852. Zeijlstra, H.H., 2007. On the syntactic flexibility of formal features. In: Biberauer, T.
Musolino, J., Crain, S., Thornton, R., 2000. Navigating negative semantic space. Ling. (Ed.), The Limits of Syntactic Variation. John Benjamins, Amsterdam, pp.
Philos. 38, 1–32. 143–174.
Musolino, J., 1998. Universal Grammar and the Acquisition of Semantic Knowledge.
University of Maryland, College Park, Unpublished doctoral dissertation.