Children do not pay attention to the language system; they have involuntary attention and memory, which means that their mind will be engaged with the semantics—the task, topic, or situation—but will not focus on the linguistic code. If they manage to acquire the form, it will be achieved indirectly, through peripheral rather than focal attention (which finds reflection in the frequent failure to acquire less salient and resilient grammatical features, visible in the structural simplification of YLs’ language, manifesting itself in the omission of grammatical functors such as auxiliary verbs, articles and bound morphemes (e.g. progressive ·ing, plural ·s and past tense ·ed); even though this may just be evidence for the Failed Functional Features hypothesis; Hawkins & Chan 1997). While children acquire their L1 when they are provided with robust, varied input and an opportunity for interaction, in order to master an L2 it is necessary to learn it consciously, especially when it is a foreign language (like English in Poland). Post-pubescent FL learners “have a conscious intention to learn the language, in contrast to young children learning their L1” (Zalewski 1995:27). Unlike children, older L2-ers, particularly adults, have the mental capacity to focus on purely formal linguistic features and engage in ‘planned language use’ by paying deliberate attention to the language forms they choose.
Felix and Weigl (1991) posit that humans have two distinct cognitive systems for dealing with the abstract system manipulation required by language. UG operates in L1A only and is accessible to very young children. Adults, on the other hand, tend to fall back on alternative, more general non-language-specific cognitive learner mechanisms of the kind described by Piaget (1923), which develop much later (around puberty) and function as a general problem solver, and they employ it in the way they would approach any field of study. They are able to make sense of even highly complex linguistic data using their general cognitive abilities: “[a]dult-acquired second languages …, however fluent they might be, could not be systems of grammatical competence” (Newmeyer 1998:74); rather, they construct surrogate systems (cf. Bley-Vroman 1989:52, his Fundamental Difference Hypothesis from a year earlier, and Section 1.5), drawing on L1 knowledge and general problem-solving mechanisms: “even supposing that the original scheme of Universal Grammar is no longer available, the foreign language learner can, in a sense, ‘reconstruct’ much of it by observing the native language” (Bley-Vroman 1988:26).
Gozdawa-Gołębiowski (2003:170f.) provides evidence for this pointing out that the notorious problems that Polish post-pubescent learners of English have in grasping the difference between subject and object questions such as (1) and (2) testify to the essentially non-linguistic nature of the mechanism they employ to make sense of the incoming data:
(1) Who does Romeo love?
(2) Who loves Romeo?
This is one argument why formal instruction will be more successful with older learners: while children are better at picking up language implicitly, adult learners perform incomparably better in activities involving analysis and memorisation. Ellis (1985a:108) concludes: “One obvious difference between the young child and the adolescent or adult is the ability of the latter to comprehend language as a formal system. Older learners can learn about language by consciously studying linguistic rules” and apply these in production. “In contrast, younger children, while not totally lacking in meta-awareness, are not so prone to respond to language form” and overt correction (ibid.) where morphosyntax is concerned (this came to be known as the ‘No Negative Evidence’ Hypothesis; cf. Section 1.11.14). For the young child—who is often unaware of the mechanisms of the learning process itself—language is but a vehicle for expressing meaning.
Children have a short attention and concentration span, while learning grammar is more like an exact science – it requires concentration. Moreover, children tend to focus their attention on the end of words and add suffixes and postpostitions before they notice the existence of and begin using prefixes and prepositions. Thus the teacher has to wait before affixes and prepositions can be taught successfully.
Even though children may be able to look at language as a separate object for exploration and to distinguish certain parts of it while disregarding others (Chaudron 1983; Karmiloff-Smith 1986, 1992), their approach to language is nevertheless holistic – they do not analyse it or break phrases into chunks, but treat and learn it formulaically, integrated with other skills, and there is a clear developmental gap between their ability to use the L1 and their metalinguistic awareness (Karmiloff-Smith 1979). In the incipient stages of LL they perceive ‘chunks’ of speech without awareness of their internal structure, but in order to achieve progress they need to be able to analyse input so as to be able to use it creatively and generate novel utterances. The young child lacks flexible thinking and can only perceive surface similarities and—at best—some deeper analogies. Learning grammar requires analytic skills, especially in the case of contrastive teaching, which necessarily entails a more ‘mentalistic’, cognitive technique with an explicit presentation of points of similarity and contrast (e.g. Rivers 1968; Stockwell 1968; Jakobovits 1969). Antonymous relations are not usually acquired by the age of five and the predisposition to recognise differences and to think flexibly—and, by extension, to deal with information pertaining to the organisation of language—does not come until the onset of abstract thinking that takes place around the age of 12-14 (cf. Matczak 1992; Dakowska 2005:29), roughly with the final stage of cognitive development as described by Jean Piaget (i.e. Formal Operations). Hence, children’s L2 production does not depend on explicit knowledge, but must be developed implicitly; “metalinguistic awareness has little or no role to play macrodevelopmentally in language acquisition” (Karmiloff-Smith 1986:139).
Children’s capacity for a conscious learning of forms and grammatical patterns is still relatively undeveloped; it is only gradually that they become able to generalise and systematise. Without the ability to infer and generate grammar rules, and to identify causal relations between various occurrences, they require constant repetition. What additionally matters for the Language Interface Model is that YLs are neither cognitively nor psychologically mature to make comparisons between the L1 and the TL, which is why grammatical explanations involving cross-linguistic associations are believed to be more beneficial for the adult learner (Titone & Danesi 1985).
Problem-solving denotes the diagnosis of what element is missing in the initial set of information (problem space), where the solution must be generated by the subject rather than found among the data available (Dakowska 2005:133). It would be unwise to posit this very demanding form of reasoning in young children, and yet this is what is necessary for hypothesis formation and testing, as well as for deduction and induction activities.
Furthermore, children cannot control what they are taught; the younger the learner, the patchier storage and recall, which again makes recycling activities necessary, whereas age improves LL capacity. Memory consists of three phases: registering, storing (based on repetition, which may be passive) and recalling (based on active repetition). In order to be able to say that we have learnt a given item successfully, all three stages must be available (actually, the learners who progress most rapidly may be adolescents, as they may have better memories than adults).
Children have limited life and learning experience. Adults, in comparison, do not enter the FL classroom as virgin tabulae rasae, but bring in a wealth of background knowledge and a long history of learning experiences on which the teacher can effectively capitalise to facilitate their learning; especially as they are all already masters of one language, frequently having the additional invaluable experience of learning another (Pratt-Johnson 2006:14).
Moreover, children’s reasoning is concrete, concerned with tangible hic et nunc realia and situations whose reference is observable in the immediate environment, ergo we can only teach to them what we can present or demonstrate. Abstract reasoning is tied to biological growth and does not develop until between 11 and 14 years of age, which is why abstract grammatical notions and patterns are beyond children’s grasp.
Language learning aptitude—a cluster of the specific cognitive qualities necessary for SLA, separate from the general ability to master academic skills—is prone to influence both the rate of development (with Carroll’s (1963) understanding of the concept as “how long the learner takes to learn a given amount of material rather than the amount of material he can learn”), notably where formal classroom learning is concerned (those learners with a gift for formal study are prone to progress more rapidly), and ultimate success in SLA, particularly if this is measured by formal tests of linguistic competence. Four major components of aptitude can be identified, nearly all reliant on the learner’s developing maturity (cf. Carroll & Sappon 1959; Stern 1983; Dörnyei 2005):
(1) phonetic coding ability– the auditory capacity to perceive, discriminate, memorise, and articulate the meaningful sounds (phonemes) of a language;
(2) grammatical sensitivity– “the individual’s ability to demonstrate awareness of the syntactic patterning of sentences in a language” (Carroll & Sappon 1959:7), of the functions and patterning of forms, and to establish correspondence between graphemic and phonemic representations;
(3) rote learning ability, to store and recall language material and establish and retain associations between sound and meaning; and
(4) inductive (language learning) ability– the ability to infer and notice both morphosyntactic and semantic regularities of the language, to identify similarities and differences, and patterns of relationship involving meaning and form.
Aptitude thus appears to be age-related and develop along with the general ability for abstract thinking.
Children are quick to learn words (they learn predominantly through mimicry, and this concerns not only language, but also all other kinds of knowledge as well as behaviour and skills), but slower to learn complex phrases and structures, which poses the necessity of a constant repetition and recycling thereof. While vocabulary is based on mechanical, short-term memory (the memory for rhyme and rhythm, which relies on frequent exposure and repetition, the earliest type of memory and therefore predominant in young children), grammar is based on logical, long-term memory – a memory for patterns, which develops very slowly (between around 11 and 14 years of age, in conjunction with abstract thinking tied to biological development) and does not reach full competence until around puberty (except dyslexic children, whose semantic memory comes first, but the mechanical one must be trained). Learners under the age of 12-13 can ably repeat and memorise long words and expressions, but are not able to analyse them as logical memory is not well developed yet.
Walsh and Diller (1981) distinguish two broad types of functioning of the left cerebral hemisphere. Lower-order functioning, associated with Wernicke’s and Broca’s areas (responsible for the understanding and production of speech respectively, also known as posterior and anterior speech cortices), involves basic syntactic processing and the motor operations employed in speaking and writing. Higher-order functioning, associated with a different area of the cerebral cortex, involves semantic processing and verbal recognition. Walsh and Diller suggest that while younger learners rely primarily on lower-order processing, which is a function of early maturing, higher-order processing is contingent upon late developing neural circuitry and is therefore available for use only in older learners.
It is also conceivable that, as Krashen (1982) speculates, inasmuch as older learners are prone to be more involved in sustaining a conversation, they will progress more rapidly than younger ones. (After all, few children display fascination with the meaning expressed through the exhaling noises produced by another person, while lengthy debates of intellectual and other nature form our daily bread.)
Into the face of the young man who sat on the terrace of the Hotel Magnifique at Cannes there had crept
a look of furtive shame, the shifty hangdog look which announces
that an Englishman is about to talk French.
—P.G. Wodehouse (1935) The Luck of the Bodkins, p. 1
While real motivation comes from within each individual (Harmer 1998:8), young learners rarely have clear motivation; they may come to class simply taking it for granted, or because they like the teacher. They will all at once be less able to assume responsibility for their learning—to use the metacognitive strategies of focusing, arranging, planning, monitoring, and evaluation (Oxford 1990:16)—in effect also ruling out any serious attempt at large-scale comparative assessment of their progress. Older, especially adult students will know the importance and see the point of study.
While adolescents and adults, when they embark on the language-learning program, have—for the most part—already developed the skills of reading or writing, children are far behind taking their first steps with the alphabet; how, then, can we expect them to be able to analyse language data, if these can only be presented as fleeting utterances, with the capacity limitations of short-term memory not allowing the learners to store more than a couple of utterances at the same time, at best? With YLs a whole new code must be taught alongside the introduction of literacy. Collier (1992) interpreted studies of bilingual children in the early grades as indicating that, for ultimate literacy and academic achievement in the L2, L1 instruction is more important than L2 instruction. This chimes in with the interdependency principle formulated by Cummins (1981b) to refer to the idea that Cognitive/Academic Language Proficiency (CALP) is common across languages, and can—therefore—easily be transferred by the learner from L1 to TL use; explaining why persons literate in their L1 face fewer problems developing CALP than ones who are not.
A great many words—if not the preponderance—refer not to individual entities, persons, events, etc., but their classes. Thus, children who are still acquiring their L1 have to establish the range of reference of the lexical items and find out the boundaries of the relevant classes (Swan 1997). While older L2-ers are also faced with this predicament, they have the great advantage over YLs in that they already know most concepts and have established how their culture pigeonholes the world.
source: Paradowski, Michał B. (2007) Exploring the L1/L2 Interface. A Study of Polish Advanced EFL Learners. Institute of English Studies, University of Warsaw, pp. 247–52.
 Cf. also Eysenck’s (1998) distinction between focused attention, where only one input is processed, and divided attention, where more are responded to.
 Although the presence of articles in the input may actually make acquisition easier by providing appropriate information with respect to the focus of the utterance and the identification of the referents.
 See also Dakowska (1993) for a cognitive psycholinguistic account of language (its learning and use) as a specialised cognitive faculty rather than an encapsulated, domain-specific module.
 Bley-Vroman cites the following characteristics which distinguish adult FLL—which bears more affinity to other kinds of general learning—from child L1A:
- characteristically frequent lack of success in FLL (vs. L1A);
- the rarity of complete success, especially with regard to subtle grammaticality judgments and pronunciation;
- variable success, as with other cognitive skills;
- variation in the goals set—as in other skills—absent from L1A;
- high negative correlation between age of arrival and ultimate proficiency attained by immigrants;
- frequent fossilisation;
- typically indeterminate intuitions suggesting incomplete competence even in advanced L2-ers;
- importance of instruction and practice.
 But they do respond to corrections of lexical mistakes, and bilingual children may have a more highly developed sensitivity to form.
 Cf. Slobin’s (1973) operating principles referring to the various learning strategies employed by children during L1A.
 Which is why they are generally happier and carefree.
 This is also why in L1A dexical words such as concrete nouns are taken to be acquired before indexical pronouns and adjectives (Knüppel et al. 2007).
 Quoted in Cook, Vivian http://homepage.ntlworld.com/vivian.c/SLA/Motivation.htm
 CALP, manifested in “the manipulation of language in decontextualized academic situations,” was seen as Cummins (1979) as one of two components of language proficiency, the other comprising basic interpersonal and communicative skills (BICS), i.e. “the manifestation of language proficiency in everyday communicative contexts.”