Language Teaching

http://journals.cambridge.org/LTA

Additional services for Language

Teaching:

Email alerts: Click here

Subscriptions: Click here
Commercial reprints: Click here
Terms of use : Click here

Error Analysis, Interlanguage and

Second Language Acquisition

S. P. Corder

Language Teaching / Volume 8 / Issue 04 / October 1975, pp 201 - 218

DOI: 10.1017/S0261444800002822, Published online: 23 December
2008

Link to this article: http://journals.cambridge.org/

abstract_S0261444800002822

How to cite this article:

S. P. Corder (1975). Error Analysis, Interlanguage and Second
Language Acquisition. Language Teaching, 8, pp 201-218
doi:10.1017/S0261444800002822

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/LTA, IP address: 138.251.14.35 on 06 Apr

2015
Survey article

ERROR ANALYSIS, INTERLANGUAGE AND

SECOND LANGUAGE ACQUISITION

S. P. Corder
University of Edinburgh

1.
In the course of learning a second language, learners regularly produce
utterances in speech and writing which judged by the rules of the second
language are erroneous, or ill-formed. Traditionally the attitude to errors was
that they were a sign that the learner had not yet mastered the rules he was
taught and that they were therefore to be dealt with by repeating the explanations
until they disappeared. If learning were efficient errors would not occur. This
point of view gave way later to the notion that errors were an indication of the
difficulties the learners had with certain aspects of the language, which could
be explained by the persistence of the habits of the mother tongue and their
transfer to the new language (Lado, 1957). In this case they were to be dealt
with not by further explanation of the target language rules but by more
intensive drilling of the sound patterns and sentence structure of the language.
Errors were the result of interference and in an ideal teaching situation could
be avoided (Lee, 1970). The difficulties of learners could be predicted by a
comparison or contrast between the structures of the mother tongue and the
target language and appropriate steps could then be taken to minimise the
difficulty and reduce the interference. From this notion has developed the whole
industry of 'cohtrastive linguistics', with research projects and regular publica-
tions of results in a number of countries. The body of literature in this field
is very large and although increasingly seen as related to the field here being
reviewed, merits separate treatment. Several bibliographies on the topic are
available (cf. Thiem, 1969). For an authoritative recent statement of the
'classical' position see Nickel (1971 a) and for a critical study of the 'state of
the art', Eliasson (1973).
In more recent years doubts have increasingly been voiced about the status
and applicability of contrastive linguistic studies to language teaching (Ritchie,
1967; Nemser, 1971; Slama-Cazacu, 1971; Dulay & Burt, 1974rf), firstly
because not all difficulties and errors can be traced back to the influence of the
mother tongue (Richards, 1971a; Dulay & Burt, 1973; Duskova, 1969) and
consequently other explanations must be sought; secondly, what contrastive

I-TA \ 4 201

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
SURVEY ARTICLE

analysis predicted as a difficulty did not always in practice turn out to be so

(Nickel, 19716); and thirdly, the purely theoretical problems of making
adequate comparisons of languages made the whole operation of doubtful
validity (Hamp, 1968; Van Buren, 1974; Krzeszowski, 1974). As a result the
theoretical basis for such studies has been questioned and its value for language
teaching reappraised. Wardhaugh (1970) makes a clear distinction between the
strong and weak hypothesis of contrastive linguistics: The strong hypothesis
states that the difficulties of the learner can be predicted by a systematic
contrastive analysis and teaching materials can then be devised to meet these
difficulties. The weak hypothesis claims no more than an explanatory role for
contrastive linguistics: where difficulties are evident from the errors made by
learners, then comparison between the mother tongue and the second language
may help to explain them. A reasoned reply to all these criticisms has been made
by Nehls (1974) and notably by James (1971 a) in which he argues that analysts
have never explicitly made many of the claims for which they are attacked.
Nevertheless there has been a gradual abandonment by contrastive analysts since
1968 of the stronger claims and, increasingly, research projects in this area have
broadened their scope in two directions: firstly towards more theoretical
objectives in language typology and the search for universals (always a preoccu-
pation of one branch of linguistic enquiry); and secondly towards psycho-
linguistic orientation concerned with the explanation of second language
acquisition. Here it merged significantly with error analysis as we shall see. This
new development of contrastive analysis has been called 'contact analysis'
(Nemser & Slama-Cazacu, 1970; Slama-Cazacu, 1971). These authors suggest
that the task of contact analysis is to 'explain and predict language learner
behaviour, with the concrete aim of developing a more scientific approach to
the processes of foreign language teaching'.

2. Error analysis
There are now a number of general statements of the ' state of the art' which
give a general comprehensive account of what error analysis is concerned with:
Nickel (1972) in German; Lange (1974) in French; and Corder (1973),
Svartvik (1973), Richards and Sampson (1974), and Schumann and Stenson
(1975) in English.
Contrastive analysis developed in a climate in linguistics and psychology
which can be broadly characterised as 'structural' and 'behaviourist'. 'Struc-
turalism ' in linguistics took the view that the structure of every language was
sui generis and therefore to be described in its own terms. Consequently it
followed logically that languages could not be compared. It was therefore
somewhat paradoxical to attempt to account for learners' difficulties, which were
clearly related to features of their mother tongue and explained psychologically

202

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
 SURVEY ARTICLE

as the transfer of their mother-tongue habits by undertaking a theoretically

impossible task. With post-structuralist developments in linguistics, associated
with the name of Chomsky, a willingness to seek common or even universal
features in human languages became again a goal of linguistics, but now
explicitly explained in psychological terms as inherent properties of the human
mind. Language acquisition and second-language learning could now be ap-
proached as a problem of cognitive learning and the possession of a second
language was seen as the possession of knowledge of a certain kind ('compe-
tence ') rather than as a set of dispositions to respond in a certain way to external
stimuli. A language user possesses a set of cognitive structures acquired by some
process of data-processing and hypothesis formation in which the making of
errors was evidence of the learning process itself and probably not only
inevitable but necessary (Dulay & Burt, 1974 d). It now became relevant to study
a learner's linguistic performance in detail in order to infer from it the nature
of that knowledge and the processes by which it was acquired. From the
insights gained from such investigations one might be able to adapt the teaching
methods and materials in order to facilitate the process of acquisition. Central
to the investigations was the analysis of the errors made by learners since they
represented the most significant data on which a reconstruction of his knowledge
of the target language could be made. This is essentially the point of view
presented by Corder (1967). He speculated that the processes of first- and
second-language acquisition were fundamentally the same and suggested that
when the utterances of first- and second-language learners differed, as clearly
they did, these differences could be accounted for by differences in maturational
development, motivation for learning and the circumstances of learning. The
learner was seen as constructing for himself a grammar of the target language
on the basis of the linguistic data in the language to which he was exposed and
the help he received from teaching. This process has been called the 'creative
construction hypothesis' by Dulay and Burt (19746) following Roger Brown
(1973). The grammar he created for himself is referred to by Nemser (1971)
as an 'approximative system'. These systems are evidently 'transient' and the
systematic nature of these systems is proved by ' the regularity of patterning of
errors in perception and production of a given target language by learners
sharing the same mother tongue'. James (1974) finds this proposal paradoxical.
' How can a system remain a system if it is in flux ?' He speaks of a fictitious
'homeostasis'. To adopt this point of view, however, is to deny the possibility
of all language description with available theoretical models. All languages are
in a state of flux. The notion of etat de langue is a necessary idealisation upon
which all linguistics is founded.
Nemser does, however, allow that stable varieties of' approximative systems'
are found, for example in immigrant speech where the learners have 'reached
a plateau' in their learning. Selinker (1972) devotes considerable space to this

•4-2 203

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
SURVEY ARTICLE

phenomenon of' fossilization' as he calls it. These approximative systems are

referred to by Corder (1971 a) as 'idiosyncratic dialects' of the target language,
a point taken up by James (1971ft) when he refers to language learning as a
process of dialect expansion and points out that he has referred to what Nemser
calls approximative systems as the phenomenon of 'interlingua'.
Selinker (1972) in an influential paper refers to this same phenomenon as
'interlanguage'. This term emphasises the structurally intermediate status of
the learner's language system between mother tongue and target language, whilst
Nemser's term, 'approximative system', emphasises the transitional and dy-
namic nature of the system. Both terms have now received wide acceptance in
the literature of error analysis and second-language learning.

3. Lapses and mistakes

Chomsky makes a distinction between what a speaker knows of his language
(competence) and how he uses it for communicative purposes (performance).
Native speakers are assumed to have a perfect knowledge of the systems of their
mother tongue, but they nevertheless produce utterances which are judged
'ungrammatical' by other native speakers. It is necessary, therefore, to make
an equivalent systematic distinction between errors, typically produced by
people who do not yet fully command some institutionalised language system
(e.g. learners or dialect speakers) and mistakes or lapses, which are failures to
utilise a known system correctly (Corder, 1971a). The native speaker is
normally capable of recognising and correcting such lapses or mistakes, which
are not the result of a deficiency in 'competence' but the result of some
neurophysiological breakdown or imperfection in the process of encoding and
articulating speech. These phenomena have been studied by linguists and
phoneticians interested in explaining the process of speech perception and
production (Boomer & Laver, 1968; Fromkin, 1973; Bierwisch, 1970). These
studies give us insights into the actual production and processing of speech but
are not relevant to the explanation of language learning. It is, however,
necessary when undertaking error analysis of learners utterances to be able to
distinguish between lapses and errors, since language learners are subject to the
same failures in their performance in the second language. Unfortunately the
important terminological distinction drawn here is not always carefully observed
by all writers.

4. Methods in error analysis

The significance of the study of learners' errors given so far has been seen in
its relevance to reconstructing the learner's 'approximative system' at any
particular stage in his learning career, rather than providing the practising

204

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
 SURVEY ARTICLE

teacher with information and insight of a practical sort in the developing of

teaching materials and classroom practices, e.g. corrective or remedial pro-
cedures. At this point we must note that error analysis can be seen to serve
two related but distinct functions (Zydatiss, 1974a; Strevens, 1969). The one,
pedagogical and 'applied' in aim, and the other, theoretical, leading to a better
understanding of second-language learning processes and strategies. More will
be said of each of these functions, but common to both aims is the need for
an adequate linguistic explanation of the nature of the errors found in any
particular learning situation. We are here concerned with the methodology of
description. Until we are able to give a linguistic account of the nature of
learners' errors we can neither propose pedagogical measures to deal with them
nor infer from them anything about the processes of learning. However, even
here the divergent aims of error analysis have a relevance. To understand the
learning process we must study the development of individual learners in
relation to their particular learning settings, social and linguistic, whilst peda-
gogical objectives are served by the study of errors in the performance of
learning groups, i.e. groups of learners, homogeneous in terms of age, sex, stage
of learning or mother tongue (Corder, 1973). This difference of objective
determines the data for analysis. It is at this point that a useful terminological
distinction has been proposed by Svartvik (1973). He suggests that the term
'error analysis' should be reserved for the study of erroneous utterances
produced by groups of learners at some stage of their learning career and
' performance analysis' for the study of the whole performance data (not just
erroneous utterances) from individual learners in the longitudinal studies called
for by Corder (1973) and Hammarberg (1973).
For pedagogical purposes we need to know what are the principal learning
difficulties of groups of learners (their well-formed utterances are assumed to
be evidence of an absence of difficulty). To achieve this we need a qualitative
linguistic classification of errors, and a quantitative statement of the relative
frequency of each type of error. We need further some evaluation of the gravity
of each type of error from a communicative or pedagogical point of view, so that
we may assign priorities to the treatment of each problem, and finally we need
some explanation of the cause of each type of error so that we undertake
appropriate remedial measures.

4.1. Classification of errors

The traditional classification into errors of omission, addition, substitution and
word order is too superficial to be of benefit to the learner or to explain
difficulties (Corder, 1972). Satisfactory classifications begin with an analysis
which assigns errors to levels of language description, i.e. errors of orthography
or phonology, of morphology or syntax, of vocabulary, and within each level

205

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
SURVEY ARTICLE

according to systems, e.g. vowel or consonant systems, tense, aspect, number,

gender or case. More recent classifications attempt to explain errors linguistically
within the framework of various generative and transformational models of
description. In such cases errors are described in terms of breaches of the rules
of the grammar or phonology. The more descriptively adequate the aims of the
classification are the more difficult the task.
In order to make a classification one mustfirstdistinguish between mistakes and
errors and determine that error is in fact present. Not all apparently well-formed
utterances are error-free (Corder 19716) nor is error-free performance any
indication that the learning goal has been reached (Levelt, forthcoming). In
order to identify the presence and nature of an error an interpretation of
the learner's utterance is necessary, but it is not always easy to know what
the learner was trying to say. Hence the central role of interpretation in the
techniques of error analysis (Corder, 1972). Many formally erroneous utterances
are potentially ambiguous even when taking the context into account. Often it
is only possible to discover the learner's communicative intent by asking him
what he means. But access to the learner is also necessary because learners do
not necessarily provide sufficient evidence to build a picture of their approxima-
tive system from their spontaneous utterances alone (Duskova, 1969). There-
fore techniques of controlled elicitation, i.e. tests which force the learner to
reveal some specific aspect of his interlanguage, are necessary (Corder, 1974a;
Zydatiss, 1974c). These controlled elicitation techniques are of various sorts:
multiple choice tests, translation tests, imitation tests (Naiman, 1974; Swain,
Dumas & Naiman, 1974). It may even be useful to get the learner to introspect
about his knowledge (Kellerman, 1974). Techniques of controlled elicitation
are however not easy to apply in the case of young children and consequently
investigations of child second-language learning have resorted to techniques
similar to those proposed by Brown (1973) for the study of first-language
acquisition. Thus, Burt, Dulay and Hernandez (1973) use a technique they have
called the bilingual syntax measure which is an instrument designed to elicit
natural speech from children, not specific responses, so that the proficiency with
which the child uses certain specific grammatical structures can be logged.

4.2. Evaluation
The need to make judgements on errors derives from two sources: firstly the
need to assess a learner's knowledge for administrative purposes (the assignment
of grades/marks) and secondly to determine priorities for remedial measures.
For a general discussion of evaluation see Nickel (1972, 1973). Attempts have
been made to use linguistic criteria for the establishment of error gravity. James
(1974) proposes an assessment based upon the number and nature of the rules
transgressed in order to measure the degree of deviance of an error from the

206

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
 SURVEY ARTICLE

correct target language form and Olsson (1972) distinguishes between syntactic
and semantic errors. An alternative linguistic evaluation is to be found in Burt
and Kiparsky (1975). They make a hierarchical distinction between globalerrors
which involve deviance in the overall structure of sentences and local errors
involving the structure of constituents of simple sentences or subordinate
clauses. Another approach to evaluation is to measure the degree of disturbance
an error may have on the efficiency of communication in terms of its frequency,
generality or comprehensibility (Johanssen, 1973), or to measure its gravity by
the degree of tolerance extended to it by native speakers or language teachers
(Lindell, 1973; Johanssen, forthcoming; James, 1975; Olsson, 1973,1974). The
problem of the relationship between the degree of linguistic deviance of an error
as determined by some sort of linguistic measure and its comprehensibility or
intolerability to native-speaking judges is still far from being understood.

4.3. Explanation
Whereas identification and description of errors is a matter for linguistics, the
explanation of errors is a matter for the psychology of second language learning:
it bridges the gap between error analysis as a pedagogical exercise and performance
analysis as part of the investigation into the processes of second-language
learning. In order to deal with errors teachers must be able to account for why
they occurred, but in the broader considerations of performance analysis, it is
the learner's whole performance which is in focus. Here no distinction is made
between 'erroneous' and 'correct' utterances. Both are evidence of the learners
approximative system at a particular stage in his interlanguage development
(Corder, 1971 b). What is of interest is how his approximative system as a whole
came to be as it is.
It is usual in error analysis to identify three principal causes for error
(Richards, 19716): language transfer (Selinker, 1969) gives rise to interlingual
errors, in which the learner's errors are accounted for by interference from the
mother tongue. George (1972) found that as many as one third of the deviant
sentences of learners could be attributable to this cause. Other workers have
found similar or greater proportion (Grauberg, 1971; Duskova, 1969). In the
case of child learners, however, as low a proportion as 3 per cent could be
ascribed to this cause (Dulay & Burt, 1973). It is clear that many factors play
a part in causing transfer errors: age of learner being the principal one, but also
the formality of the learning situation and the method of teaching.
A second class of error has been called 'intralingual' by Richards (19716).
These errors do not reflect features of the mother tongue, but result from the
learning process itself. Learners are seen to make inductive generalisations about
the target language system on the basis of the data to which they are exposed.
Since the data is necessarily restricted they will tend to overgeneralise and

207

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
SURVEY ARTICLE

produce incorrect forms by analogy. This type of error has been well documented
by Jain (1974). The result of this process is to reduce the target language system
to an apparently 'simpler' form (Richards, 19746) or more 'regular' system
(Slama-Cazacu, 1974). These types of error may also be regarded as develop-
mental (Dulay & Burt, 1973), since similar processes are regularly observed in
child language acquisition studies. Since errors having this provenance are
independent of the mother tongue of the learner, one will find that they are
common to all learners of any given second language. This provides some
theoretical validity to the collections of' common errors' which have been made
from time to time (Fitikides, 1967; French, 1949; Ballard, 1970).
A third source of error is assigned to faulty teaching techniques or materials.
Richards (1971a) calls this process 'hypothesizing false concepts'. Little
systematic study of this cause of error has been made and, clearly, errors not
readily classed as inter- or intra-lingual cannot be confidently assigned for this
reason to this third category. Only prolonged observation of sets of learners in
the learning situation permits the identification of such a cause of error. A recent
study by Stenson (1975) is the fullest account available so far of what she calls
'induced errors'. Corder (1973) regards this class of error as the only redundant
error from a language learning point of view. This same source of error is
accounted for, somewhat idiosyncratically, by Selinker (1972) under the rubric
of 'transfer of training', not to be confused with language transfer as a source
of error. A recent interesting analysis of error from a psychological 'skill
theory' approach with pedagogical implications is that of Levelt (forthcoming).
The number of published and unpublished descriptions of errors found in
corpuses drawn from different teaching situations is very large. The most recent
comprehensive bibliography of works on error analysis and related topics
(Valdman, 1975) lists no less than 246 separate case studies, the majority of
these having been produced in the last ten years and carried out in the light
of the sort of theoretical orientations already outlined. They cover principally
errors made by learners of English, French, German and Spanish from varying
linguistic backgrounds and tend to deal separately with errors of syntax,
pronunciation, spelling and vocabulary.

5. Performance analysis
A distinction has been drawn between studies of errors with a pedagogical
objective of pointing to the development of appropriate remedial techniques and
materials, and performance analysis, the study of the learner's language system
in order to discover the psychological processes of second language learning.
Selinker (1972) postulates five central processes, as he calls them, which are
effective in determining the nature of a learner's interlanguage: language
transfer, already discussed, transfer of training, i.e. teaching induced (incorrect)

208

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
 SURVEY ARTICLE

hypotheses about the target language, strategies of second language learning, the
learning strategies of the individual learner, which leads him to purely idiosyn-
cratic hypotheses about the target language, such as a preference for holistic
learning rather than an inductive analytic approach (cf. Cancino, Rosansky &
Schumann, 1974). Hatch (1974) contrasts such learners as 'data-gatherers' or
'rule-formers'. The fifth process Selinker identifies is strategies of second
language communication. Strictly speaking, this is not a process of second-language
learning, but is invoked to account for the phenomenon of fossilisation. This
is a state of affairs when the learner ceases to elaborate or 'complexify' his
approximative system in some respect, however long he is exposed to new data
or new teaching. Selinker notes the important fact that learners regularly regress
to an earlier approximative system under particular circumstances of communi-
cative need.
The study of strategies of communication has been taken further by other
investigators. The notion is that second-language learners adopt certain iden-
tifiable strategies when faced with the need to communicate with a less than
adequate interlanguage system. Levenston (1971) identifies such features as
over-formality or over-informality, verbosity, use of substandard forms, under-
differentiation, interchangeably, archaism as the unintended stylistic results of
this inadequacy. Varadi (1973) compares the same messages in the mother
tongue and the interlanguage of the learner to investigate how the learner copes
with his communication problem, whilst Richards (1972) points to simplification
as the result of such communication strategies and draws attention to the
important structural similarities between pidgins and interlanguage: Linnarud
(1975) compares the lexical usages of learners and native speakers in similar
communication tasks, whilst Zydatiss (1973) investigates the availability of
various 'thematising processes' in the interlanguage of German learners of
English. Widdowson (forthcoming) treats the learner's performance as evidence
of the possession of idiosyncratic ' rules of expression' in his ' communicative
competence' in the target language. There is still a great deal of work to be done
in understanding how learners use their interlanguage to achieve their com-
municative purposes.

6. Second-language acquisition studies

In section 2 an account was given of the 'creative construction' hypothesis of
second-language learning, which proposed that the learner created for himself
a series of hypotheses about the grammar of the target language by some
process of data-processing, hypothesis formation and testing. In this theory all
utterances made by learners serve to test his hypothesis and his errors are
evidence of false hypotheses. The making of errors is a necessary part of
language learning. As Dulay and Burt (1974rf) say, 'You can't learn without

209

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
SURVEY ARTICLE

goofing.' In their important series of investigations into child second-language

acquisition (Dulay & Burt, 1972, 1973, 1974a, b, c) they classify the errors of
child learners into four categories: interference goofs, mother-tongue develop-
mental goofs, ambiguous goofs and unique goofs. This categorisation is readily
relatable to the classifications already reported above, but what is significant is
that they are able to show that the distribution and nature of these errors is
substantially the same as that of children acquiring their mother tongue. The
only difference is a very small proportion of interference phenomena. This leads
them to postulate the L 1 = L 2 hypothesis, which states that the learning of
a second language is fundamentally the same process, at least in children, as
the learning of a first language and that sequential development of the approxi-
mative systems is substantially the same in both cases whatever the mother
tongue of the learner. Such research has naturally led to a revival of interest
in the relationship between first- and second-language learning and to its
relevance for language pedagogy (Cook, 1969, 1973; Ervin-Tripp, 1970;
Newmark & Reibel, 1969; Taylor, 1974). Ervin-Tripp (1974) suggests from
her investigations that older children may regress to processing strategies
similar to those in first-language acquisition when faced with data in a second
language. Hence the occurrence of utterances bearing no superficial resemblance
to either the first or the second language.
These studies suggest that there may be some universal processing strategies
of language learning which lead to a similar natural sequence of approximative
system in all child second-language learners. These hypotheses have been
carefully examined by Hatch (1974) in the light of the now substantial body
of longitudinal studies of child second-language learners, unfortunately mostly
unpublished. For published accounts see Canciao et al. 1974; Ravem, 1971;
Dato, 1972; Dulay and Burt, 1974a; Hakuta, 1974; Milon, 1974. For a critical
discussion of the work of Dulay and Burt, see Tarone (1974).
These studies have largely been on children, and the circumstances in which
they have learned the second language have generally been informal. The
question which still has to be resolved is the influence of the learning setting,
the nature of language data and the communication functions of the second
language on the learning process. Does this influence the' natural' sequence and
nature of the approximative systems which are developed ? It is clear that if there
is any truth in the speculation (Corder, 1967) that second-language learners
might have 'built-in syllabuses' for learning second languages, as infants appear
to be 'programmed' to learn their mother tongue (Brown, 1973), such
knowledge would be of immense value for the design and sequencing of
instructional materials.

210
http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
 SURVEY ARTICLE

7. Simplified linguistic systems

The discussion of the learner's language system so far has been in the light of
linguistic and psycholinguistic notions. Language is a means whereby human
beings communicate with each other in a social setting. The nature of the
learner's approximative system as he constructs it must be sensitive to the needs
he feels to be able to use it for communicative purposes. As Schumann (1974)
points out, the need to elaborate and develop his approximative system is a
response to the learner's widening communicative needs. When these are
restricted, as they are when the learner does not seek integration into the target
language community, a simpler version of the target language system emerges
(Richards, 1972). This interlanguage system is seen to be structurally simpler
by comparison with the fully complex target model. It has now become evident
that interlanguage systems may often bear considerable resemblances to other
simple linguistic codes such as pidgins and Creoles, which are now recognised
to bear strong resemblances to each other whatever their linguistic provenance.
This is particularly true when the interlanguage systems have developed in
informal learning settings. Clyne (1968) noted, for example, the strong resem-
blances of the interlanguage of immigrant workers in German whatever their
mother tongues, and the similarity of these to pidgins. It is now generally agreed
that the structural simplicity of pidgins is a result of their limited communicative
function; development into a Creole is a result of the broadening functional needs
of their speakers. Thus the process of elaboration (creolisation) occurs when
the need to express social identity and affective functions is felt. The same needs
may account for the development of the mother tongue in infants. For this reason
the study of pidginization and creolisation may yield valuable insights into the
second-language learning process. Valdman (forthcoming) points out that
creolists have recently shifted their interest towards the social-psychological
processes underlying the development of pidgins and Creoles and that this has
brought these studies closer to the interests of those concerned with second
language learning processes. Richards (19746) suggests that simplification is
perhaps a 'universal learning strategy'. We have already noted the process of
overgeneralisation in learning and seen that this results in approximative
systems which are structurally simpler than those of the target language.
Samarin (1971) has even posited that there may be 'universal rules of simplifi-
cation '. However, as Corder (forthcoming) has pointed out, we cannot simplify
what we do not possess and a language learner can scarcely be said to be
simplifying the rules of the target language in any psychological sense. What
results from his learning strategies may however result in a system which is
linguistically simpler.
lnterlanguages and pidgins are not the only simple codes which we meet.
Ferguson (1971) draws attention to the fact that language communities regularly

211

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
SURVEY ARTICLE

use institutionalised and stereotyped codes for special communicative functions

which are also simple in the same sense, and that these codes resemble,
structurally, interlanguage systems. Such reduced codes as foreigner-talk
(Ferguson, 1975) or baby-talk (Ferguson, 1964) are found in many language
communities. Corder (forthcoming) has speculated that we do not in fact learn
these codes anew as separate systems in later life but that they represent part
of every native speaker's competence and represent, as it were, fossilised stages
in the progressive elaboration of his knowledge of his mother tongue, available
to him for use on socially approved occasions. Thus, in a sense, the whole of
a person's linguistic development remains available to him all his life, and is
thus part of his knowledge of language when he comes to learn a second
language. The starting point of learning a second language is thus not necessarily
the fully complex adult code of his mother tongue but might be one of its
simpler codes. Widdowson (forthcoming) makes a similar point when he says
that' simplification is a process whereby a language user adjusts his behaviour
in the interests of communicative effectiveness'. If this is true of the fully mature
native speaker, then it may well be true of learners and account for the apparent
regressions to earlier (simpler) approximative systems under the pressure of
communicative needs noted by Selinker (1972). As Valdman (forthcoming) has
pointed out, this throws doubt on the whole validity of attempting to reconstruct
a learner's approximative system on the basis of his recorded utterances and
indeed on the usefulness of the systematic distinction between competence and
performance insisted upon by Chomsky.

8. Pedagogical implications of error and performance analysis

Indications of the pedagogical relevance of the studies discussed above have been
made throughout the text. They fall into three categories: the problem of
correction; the design of syllabuses and remedial programmes; and the writing
of pedagogical grammars. All these are related to those studies which I have
called error analysis. It is too early to draw any conclusions of an immediate
practical sort from the work proceeding in performance analysis and the study
of second-language learning processes, though some hints of what these may
be like have been made, cf. Widdowson (forthcoming). For a general discussion
of the relevance of errors to language teaching see Gorbet (1974). The problem
of correction is two-fold: what to correct and how to correct. The first question
relates to the assessment of the gravity of the error in terms of its interference
with comprehensibility or the degree of linguistic deviance. George (1972)
considers that sheer frequency is no criterion for decisions about whether to
correct, whilst Sternglass (1974) maintains that comprehensibility of an incor-
rect utterance is not a reason for not correcting it; the degree of linguistic deviance
is also an important consideration. Holley and King (1975) stress the need to

212

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
 SURVEY ARTICLE

encourage learners to communicate and to devise correction techniques with this

always in mind, for example by requesting rephrasing or amplifying the
message, in the way that adults react to infants' utterances in their mother
tongue. Burt and Kiparsky (1975) stress that it is the global mistakes, in their
analysis, which are unique to second-language learners and cause maximum
interference with communication and hence demand attention. Local errors
will to a large extent look after themselves. A similar proposal is made by Burt
(1975).
The relevance of performance analysis to the designing of syllabuses is based
on the notion that there is some 'natural' sequence of elaboration of the
approximative system of the second-language learner and that when/if this can
be well established it would provide a psychological logic to the ordering of
material in a syllabus (Corder, 1967). Nickel (1973) flies a controversial kite
when he suggests that the language-teaching materials should reflect the
sequence of approximative systems of the learner to the point of actually
teaching 'incorrect forms': 'If intralingual steps in a system of etats de dialecte
are necessary steps within the process of language acquisition, then one would
in theory even have to consider the problems of error being built into the
language material of any kind whatsoever.' This is at least the principle which
lies behind the teaching of ITA. Up till now little experimental"work has been
done in actually trying out teaching sequences in the light of error analyses.
An exception is Valdman (1975 a), who reports on a pilot study of teaching a
course in French interrogatives based upon error analysis. He too is sympathetic
to Nickel's proposals and is prepared at least to teach forms which offend the
prescriptive and linguistic purist.
As far as the design of pedagogical grammars is concerned, the effectiveness
of the presentation and practising of linguistic materials must ultimately depend
upon what is discovered about the actual processes and strategies of language
learning, that is, on performance analysis. For further discussions of this
problem, see Allen (1973). Meanwhile there do exist pedagogic grammars based
upon error analysis. Probably the foremost amongst these is Burt and Kiparsky
(1972) for English. This is designed for remedial teaching. Another excellent
example of such a grammar is Zydatiss (19746).

Bibliography
Allen, J. P. B. (1973). Applied grammatical models in a remedial English syllabus. In
S. Pit Corder & E. Roulet (eds.), Theoretical linguistic models in applied linguistics.
AIMAV, Brussels; Didier, Paris.
Ballard, V. S. (1970). Frequent errors in French. TBull, 5, 2, 1-6.
Bierwisch, M. (1970). Fehlerlinguistik. Linguistic Inquiry, 1, 397-414.
Boomer, D. S. & Laver, J. D. M. (1968). Slips of the tongue. British Journal of Disorders
of Communication, 3, 1-12.

213

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
SURVEY ARTICLE

Brown, R. W. (1973). Development of the first language in the human species. American
Psychologist, Feb. 73, 97-106.
Burt, M. K. (1975). Error analysis in the adult EFL classroom. TESOL Quarterly, 9,
1, 53-65.
Burt, M., Dulay, H. & Hernandez, E. (1973). Bilingual syntax measure. New York:
Harcourt Brace Jovanovich.
Burt, M. & Kiparsky, C. (1972). The gooficon. Rowley: Newbury House.
Burt, M. & Kiparsky, C. (1975). Global and local mistakes. In Schumann & Stenson,
New frontiers.
Cancino, H., Rosansky, E. J. & Schumann, J. (1974). Testing hypotheses about second
language acquisition: the copula and negative in three subjects. Working Papers on
Bilingualism, no. 3. Ontario Institute for Studies in Education.
Clyne, M. (1968). Zum Pidgin-Deutsch der Gastarbeiter. Zeitschrift fur Mundartfors-
chung, 35, 130-9.
Cook, V. J. (1969). The analogy between first and second language learning. IRAL, 7,
207-16.
Cook, V. J. (1973). The comparison of language development in native children and
foreign adults. IRAL, 11, 13-28.
Corder, S. P. (1967). The significance of learners' errors. IRAL, 5, 161-70.
Corder, S. P. (1971a). Idiosyncratic dialects and error analysis. IRAL, 9, 147-60.
Corder, S. P. (19716). Describing the language learner's language. CILT Reports and
Papers, no. 6, 57-64.
Corder, S. P. (1972). Die Rolle der Interpretation beiderUntersuchungvonSchulfehlern.
In G. Nickel, Fehlerkunde. Berlin.
Corder, S. P. (1973). Introducing applied linguistics. Harmondsworth: Penguin Education.
Corder, S. P. (1974 a). The elicitation of interlanguage. Special issue of IRAL on the
occasion of Bertil Malmberg's 60th birthday. Stuttgart: Julius Groos.
Corder, S. P. (forthcoming). Simple codes and the source of the second language
learner's initial heuristic hypothesis. In Corder & Roulet (eds.), Linguistic approaches.
Corder, S. P. & Roulet, E. (eds.) (forthcoming). Linguistic approaches to applied linguis-
tics. Papers from the IVth Neuchatel Colloquium, 1975. AIMAV, Brussels; Didier,
Paris.
Dulay, H. C. & Burt, M. K. (1972). Goofing: an indicator of children's second language
learning strategies. Language Learning, 22, 235-51.
Dato, D. P. (1972). The development of the Spanish verb phrase in children's second
language learning. In Pimsleur & Quinn (eds.), The psychology of second language
learning. Cambridge University Press.
Dulay, H. C. & Burt, M. K. (1973). Should we teach children syntax? Language
Learning, 23, 2, 245-58.
Dulay, H- C. & Burt, M. K. (1974 a). Natural sequences in child second language
acquisition. Language Learning, 24, 37-53.
Dulay, H. C. & Burt, M. K. (19746). A new perspective on the creative construction
hypothesis in child second language acquisition. Working Papers on Bilingualism, 4,71-98.
Ontario Institute for Studies in Education.
Dulay, H. C. & Burt, M. K. (1974c). Errors and strategies in child language acquisition.
TESOL Quarterly, 8, 129-36.
Dulay, H. C. & Burt, M. K. (1974a1). You can't learn without goofi'ng: an analysis of
children's second language errors. In J. C. Richards (ed.), Error analysis.
Duskova, L. (1969). On sources of errors in foreign languages. IRAL, 7, 11-36.
Eliasson, S. (1973). Review of Filipovic (ed.), Zagreb conference on English contrastive
projects. Studia Linguistica, 27.

214

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
 SURVEY ARTICLE

Ervin-Tripp, S. (1970). Structure and process in language acquisition. Georgetown Mono-

graph Series on Languages and Linguistics, no. 23, 313-53. 21st Annual Round Table.
Ervin-Tripp, S. (1974). Is second language learning like the first? TESOL Quarterly
8, 111-29.
Ferguson, C. A. (1964). Baby-talk in six languages. In Gumperz & Hymes (eds.), The
ethnography of communication, 103-4. Special number of American Anthropologist, 66,
6, Pt. 2. Washington, American Anthropological Assoc.
Ferguson, C. A. (1971). Absence of the copula and the notion simplicity. In D. Hymes
(ed.), Pidginization and creolisation of languages. London: Cambridge University Press.
Ferguson, C. A. (197S). Towards a characterisation of English foreigner talk. Anthropo-
logical Linguistics, 17, 1-14.
Fitikides, T. J. (1967). Common mistakes in English. London: Longman.
French, F. G. (1949). Common errors in English: their cause, prevention and cure. London:
Oxford University Press.
Fromkin, V. A. (1973). Slips of the tongue. Scientific American, 229, 110-17.
George, H. V. (1972). Common errors in language learning. Rowley: Newbury House.
Gorbet, F. (1974). Error analysis: what teachers can do. In R. Shiu (ed.), Errors: a
new perspective. Toronto: Research Division, Directorate of Studies, Civil Service
Commission.
Grauberg, W. (1971). An error analysis in German of first year university students. In
G. Perren & J. Trim (eds.), Papers for the second international congress of applied
linguistics. London: Cambridge University Press.
Hakuta, Kenji (1974). Preliminary report on the development of grammatical morphemes
in a Japanese girl learning English as a second language. Working Papers in Bilingualism,
3, 18—43. Toronto: Ontario Institute for Studies in Education.
Hammarberg, B. (1973). The insufficiency of error analysis. In J. Svartvik (ed.), Errata.
Hamp, E. P. (1965). What a contrastive grammar is not, if it is. Georgetown Round Table
Monograph, no. 21.
Hatch, E. R. (1972). Some studies in language learning. Working papers in TESL, no.
6, 29-36. Los Angeles: University of Los Angeles.
Hatch, E. R. (1974). Second language learning - universals? Working Papers in Bi-
lingualism, no. 3, June 1974. Toronto: Ontario Institute for Studies in Education.
Holley, F. & King, J. M. (1975). Imitation and correction in foreign language learning.
In Schumann & Stenson, New frontiers.
Jain, M. P. (1974). Error analysis: source, cause and significance. InJ. C. Richards (ed.),
Error analysis.
James, C. (1971 a). The exculpation of contrastive linguistics. In G. Nickel (ed.), Papers
in contrastive linguistics. Cambridge University Press.
James, C. (1971 b). Foreign language learning by dialect expansion. Paper read to PAKS
Symposium, Stuttgart, 1971.
James, C. (1974). Linguistic measures for error gravity. AVLA Journal, 12, 1, 3-9.
James, C. (1975). Judgements of error gravities. ELS, July 1975.
Johansson, S. (1973). The identification and evaluation of error in foreign languages; a
functional approach. In J. Svartvik (ed.), Errata.
Johansson, S. (forthcoming). Problems in the study of the communicative effect of
learner's errors. In Corder & Roulet (eds.) (forthcoming), Linguistic approaches.
Krzeszowski, T. P. (1974). Contrastive generative grammar: theoretical foundations. Lodz:
University Press.
Kellerman, E. (1974). Elicitation, lateralization and error analysis. York Papers in
Linguistics, 4. University of York.
Lado, R. (1957). Linguistics across cultures. Ann Arbor: University of Michigan.

215

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
SURVEY ARTICLE

Lange, M. (1974). L'analyse des erreurs: etat actuel de la recherche. In R. Shiu (ed.),
Errors: a new perspective. Toronto: Research Division, Directorate of Studies, Civil
Service Commission.
Lee, W. R. (1970). The Dolphin English Course- Teachers' companion. London: OUP.
Leisi, E. (1972). Theoretische Grundlagen der Fehlerbewertung. In G. Nickel (ed.),
Fehlerkunde.
Levelt, W. S. M. (forthcoming). Skill theory and language teaching. In Corder & Roulet
(eds.) (forthcoming), Linguistic approaches.
Levenston, E. A. (1971). Over-indulgence and under-representation - aspects of mother
tongue interference. In G. Nickel (ed.), Contrastive linguistics. London: Cambridge
University Press.
Lindell, E. (1973). The four pillars: on the goals of foreign language teaching. In
J. Svartvik (ed.), Errata.
Linnarud, M. (1975). Lexis in free production. An analysis of the lexical texture of
Swedish students' written work. Report No. 6. Department of English, Lund
University.
Milon, J. P. (1974). The development of negation in English by a second language
learner. TESOL Quarterly, 8, 137-45.
Naiman, Neil. (1974). The use of elicited imitation in second language acquisition
research. Working Papers in Bilingualism, no. 2. Toronto: Ontario Institute for Studies
in Education.
Nehls, Dietrich (1974). Fehleranalyse versus Kontrastive Analyse. Reports of 5th Annual
Congress of G.A.L. (1973 IRAL Sonderband). Stuttgart: Julius Groos.
Nemser,W. (1971). Approximative systemsof foreign language learners. IRAL,9,115-23.
Nemser, W. & Slama-Cazacu, T. (1970). Contact analysis: a psycholinguistic approach.
Revue Romaine de Linguistique, 15, 2.
Newmark, L. & Reibel, D. (1967). Necessity and sufficiency in language learning. IRAL,
6,3.
Nickel, G. (1971a). Contrastive linguistics and foreign language teaching. I n G . Nickel (ed.),
Papers in contrastive linguistics. London: CUP.
Nickel, G. (19716). Problems of learners' difficulties in foreign language acquisition.
IRAL, 9, 3, 219-27.
Nickel, G. (1972). Grundsatzliches zur Fehleranalyse & Fehlerbewertung. In G. Nickel
(ed.), Fehlerkunde.
Nickel, G. (ed.) (1972). Fehlerkunde. Berlin: Cornelsen, Velhagen & Klasing.
Nickel, G. (1973). Aspects of error analysis and grading. In J. Svartvik (ed.), Errata.
Olsson, M. (1972). Intelligibility: a study of errors and their importance. The GUME
project. Goteborg: Pedagogiska Institutionen Lararhogskolan.
Olsson, M. (1973). The effects of different types of errors in the communication
situation. In J. Svartvik (ed.), Errata.
Olsson, M. (1974). A study of errors: frequency, origins and effects. Goteborg:
Pedagogiska Institutionen Lararhogskolan.
Ravem, R. (1974). WH-questions in first and second language learners. In J. C. Richards
(ed.), Error analysis.
Richards, J. C. (1971 a). A non-contrastive approach to error analysis. ELT, 25, 204-19.
Richards, J. C. (1971 b). Error analysis and second language strategies. Language Science,
17, 12-22.
Richards, J. C. (1972). Social factors, interlanguage and language learning. Language
Learning, 22, 2, 159-88.
Richards, J. C. (ed.) (1974a). Error analysis. Perspectives on second language acquisition.
London: Longman.

216

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
 SURVEY ARTICLE

Richards, J. C. (19746). Simplification: a strategy in the adult acquisition of a foreign

language: an example from Indonesian/Malay. Paper given at the Second Conference
on the Standardisation of Asian Languages. Manila, 1974.
Richards, J. C. & Sampson, G. P. (1974). The study of learner English. In J. C.
Richards (ed.), Error analysis.
Ritchie, W. (1967). Some implication of generative grammar for the construction of
courses in English as a foreign language. Language Learning, 17, 4S-68, 111-32.
Samarin, W. J. (1971). Salient and substantive pidginization. In D. Hymes (ed.),
Pidginization and creolisation of languages. London: Cambridge University Press.
Schumann, John (1974). The implications of interlanguage, pidginization and creoliza-
tion for the study of adult second language acquisition. TESOL Quarterly, 8, 2.
Schumann, J. & Stenson, N. (1975). New frontiers in second language learning. Rowley:
Newbury House.
Selinker, Larry (1969). Language transfer. General Linguistics, 9, 67-92.
Selinker, Larry (1972). Interlanguage. IRAL, 10, 3, 219-31.
Slama-Cazacu, T. (1971). Psycholinguistics and contrastive studies. Zagreb Conference
on English Contrastive Projects. Zagreb: Institute of Linguistics.
Slama-Cazacu, T. (1974). Theoretical interpretation and methodological consequences
of 'regularisation'. In Further developments in contrastive studies. Romanian-English
Contrastive Analysis Project. Bucharest University Press.
Sternglass, M. S. (1974). Close similarities in dialect features of black and white college
students in remedial composition classes. TESOL Quarterly, 8, 3, 271-83.
Stenson, Nancy (1975). Induced errors. In Schumann & Stenson (eds.), New frontiers
in second language learning.
Strevens, Peter, (1969). Two ways of looking at error analysis. Paper given at GAL
meeting, Stuttgart (1969). Zielsprache Deutsch, 1 (1971), 1-6.
Svartvik, J. (1973). Introduction to Svartvik, J. (ed.), Errata: papers in error analysis.
Svartvik, J. (1973). Errata: papers in error analysis. Lund: Gleerup.
Swain, M., Dumas, G. & Naiman, N. (1974). Alternatives to spontaneous speech:
elicited translation and imitation as indicators of second language competence. Working
Papers in Bilingualism, no. 3. Toronto: Ontario Institute for Studies in Education.
Taylor, B. P. (1974). Toward a theory of language acquisition. Language Learning, 24,
1, 23-35.
Thiem, R. (1969). Bibliographyof contrastive linguistics. PAKS- Arbeitsbericht,2,79-96;
3/4, 93-120.
Tarone, Elaine (1974). A discussion on the Dulay and Burt studies. Working Papers on
Bilingualism, no. 4. Toronto: Ontario Institute for Studies in Education.
Valdman, Albert & Phillips, J. S. (forthcoming). Pidginization, creolisation and the
elaboration of learner systems. In Corder & Roulet (eds.) (forthcoming), Linguistic
approaches.
Valdman, Albert (1975 a). Error analysis and pedagogical ordering. In Corder & Roulet
(eds.), Linguistic insights in applied linguistics. Brussels, AIMAV; Paris, Didier.
Valdman, Albert (1975 ft). Bibliography of language learners' approximative systems and error
analysis. Washington: Centre for Applied Linguistics.
Van Buren, P. (1974). Contrastive analysis. In J. P. B. Allen & S. P. Corder, Techniques
in applied linguistics. London: Oxford University Press.
Varadi, Tamas (1973). Strategies of target language learner communications: message
adjustment. Paper given at the 6th Conference of the Rumanian-English Linguistic
Project, Timisoara, 1973.
Wardhaugh, G. (1970). The contrastive analysis hypothesis. TESOL Quarterly, 4, 2.
Reprinted in Schumann & Stenson (eds.), New frontiers.

LTAV11I 15 217

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35
SURVEY ARTICLE

Widdowson, H. W. (forthcoming). The significance of simplification. In Corder & Roulet

(eds.) (forthcoming), Linguistic approaches.
Zydatiss, W. (1973). Fehler in der Englischen Satzgliedfolge. IRAL, 11, 4, 319-55.
Zydatiss, W. (1974a). A 'kiss of life' for the notion of error. IRAL, 12, 3, 234-7.
Zydatiss, W. (19746). Lernprobleme im Englischen und ihre Behandlung in einer
didaktischen Grammatik. Linguistik und Didaktic, 20, 5.
Zydatiss, W. (1974c). Some Test Formats for Elicitation Procedures. IRAL, 12, 281-7.

218

http://journals.cambridge.org
Downloaded: 06 Apr 2015 IP address: 138.251.14.35