Cntrastive Analysis, Error Analysis
In order to examine any apsect of the learner's interlanguage  satisfactorily, the investigator usually makes  use of two different yet cmplementary techniques, namely contrastive  analysis and error analysis.

 The ultimate objective analysis is to shed light on the source of error caused by negative transfer of L1 structures and other types of errors will not be predicted  or explained by CA.error analysis, on the other hand, is a technique employed by investigators to probe the learner;s interlanguage in terms of his failure to internalize  the structures of the foreign language as well as in terms of  his achievement or progress in that language.
                                                                         
 Thus, error analysis can account for all types of errors-interlingual or otherwise, and its heavy reliance on CA in identifying the source of interlingual or transfer errors  becomes obvious .It follows then that CA is an important aspect of; and coplementary to; error analysis. It seems fair to suggest that the domain of EA is much wider than that of CA if we take into account the learner's interlanguage, the strategies s/he may use and psycholinguistic factors behind the making of errors.

  Supporters of CA hypothesis expressed their views in two different forms. Some claimed that  nearly all the difficulties that the language learner might face are the result of his first language habits and consequently it would be possible to predict the learner's transfer errors. Others believe  that not all the learner's errors  could be traceable to his/her first language and that CA explains rather than predict the learner's errors.
             
 At any rate, there is abundant evidence from research on error analysis and second language acquisition that most errors made by foreign language learners are not due to, in fact, caused by interference from the mother tongue. As such, it has been proved that very few Arab students of English make errors in the relative positions of adjective and nouns in English, since adjectives in English are usually  placed before  the head of nomimal group, where as adjs in Arabic  are post–nominal The same argument holds true for the agent–verb sequence. Corder points out:
                                                                  
Particularly in the early stages of learning a language, the learner's interlangusge frequently shows in its grammar and phonology certain characteristics of features which can be related to his mother tongue.         

Concluding remarks                                                     
As a matter of fact, transfer of L1 features into L2 varies from one linguistic level to another.  It seems that negative transfer is noticeably strong at the phonological  level and less so at the syntactic and morphological ones.


The Error Analysis Theory
One of the main contributions that error analysis is capable of making is the examination of the learner's interlanguage  in terms of what the learner has achieved and what s/he is yet to achieve.
                                                                
   Lapses, mistakes and errors                          
Corder has emphasized the point that the problem of determining what is a learner's mistake and what is a learner's error is of some difficulty and involves a much more sophisticated study and analysis of errors than is usually accorded them.  This occurs in the case of the learner's absence. An error is, thus, understood to mean any deviation from the norms of the target language on the grammatical and lexical level. Mistakes or lapses are performance errors. They are unsystematic and not reflect the learner's lack of knowledge.

  As has been mentioned earlier, EA always makes use of CA in explaining transfer errors. More important is perhaps  the point that the error analysis theory can explain the errors of a non- contrastive nature, which contribute the majority of learners' errors.

Stages involved in the analysis of errors  
EA necessarily involves distinct but related stages, which are recognition, correction, or reformulation , classification description and finally explanation.

1. Recognition
When we say that  a particular string of any length is erroneous, a decision of some kind is being made as to the unacceptability of the string in question i.e. the string not accepted by the majority of native speakers.

  The error may be due to ungrammaticality i.e. the erroneous string is either syntactically or morphologically ill-formed, which means that a syntactic or morphological rule is broken:

*Modernism is in itself…to the all movements(to all the movements) syntactic error
*The society achieve (society achieves) progress in many field. (fields) morphological error

   Unacceptability may also be due to lexical deviance by which is meant that a stretch of language does not conform to the lexical patterning of the language under consideration. For instance, the string *… great black clouds gathered here and there in the pure sky (clear) is lexically erroneous, since pure  does not collocate with sky.  

2. Correction
In this stage, the researcher is invited to suggest the  reformulation of the deviant strings in question bearing in mind the meaning intended by the learner and the context in which the deviant strings occur.  The point that deserves to be mentioned here is the possibility that a given string might be corrected in more than one way. Let's have a look at following examples:

* John wrote a book revealed in it …
* Business letter deals with business affairs.

We notice that the generic reference is maintained in the above-mentioned sentences.

3. Classification
In this stage, the teacher has to determine the type or the category of the error made. This depends largely on the way we reformulate the erroneous string.  One of the following students made the following error: * it is the basic of our world…  . This shows a grammatical error in that the headword of the nominal group  ' the basic … of our world' is omitted.  The error could equally be regarded as lexical, involving wrong selection of a lexical item: the word basic could have been mistaken for basis. Alternatively, the student might have thought that basic is the singular form of basics.  

4. Description
Description of errors requires the researcher to say what went wrong and which grammatical rule, for example, has been broken or which lexical relation has not been maintained. In fact, description of errors is a linguistic problem.

5. Explanation
This process is a psycholinguistic one; it involves knowledge of the psycholinguistic factors behind the error.
 
Reference:

A lecture delivered by the late Dr. Ali Radwan in Aleppo University for diploma linguistic students in 1993.