The CA focuses on the comparison between the learner’s native language (L1) and the target language (L2) to predict possible difficulties in learning. This approach involves systematic comparisons based on the differences between languages. From this comparison, the concept of Transference arises. Language learners tend to transfer the rules from L1 to L2, which is when errors occur.

EA, on the other hand, focuses on the actual errors learners make when using L2. By collecting and analyzing these samples, the EA understands the learning process and the reasons for making them.

Another difference between CA and EA is that the first one assumes that by knowing both languages (L1 and L2) and their differences and similarities, we can predict and thus eradicate or avoid possible errors. On the other hand, EA assumes that errors are a natural part of the language-learning process; as Corder´s article “The significance of learner´s errors” states: “…errors as sources of insight into the learning processes”.

Some of the main objections presented against EA are:

Lack of positive data: an approach that is purely error-focused analysis may overlook correct uses of the target language. Learners often use the language correctly for a communicative purpose despite making errors.

Sometimes students do not dare to use difficult structures and stick to well-known ones. This strategy does not let the teacher/investigator realize if the learner has acquired the structure correctly.

EA does not take into account other factors such as psychological, environmental or emotional ones that a learner may be going through.

Some of the main objections against EA are that it is not clear how to identify which errors are caused by lack of knowledge and which are caused by random factors. It is also not clear because correct and incorrect forms can appear together. And, even though native speakers may feel confident about what sounds right or wrong in their language (native speaker intuitions), it is still difficult to define what counts as an error. Sometimes, what seems like a mistake to one person might be acceptable to someone else, and there are also other factors like the situation, whether they are writing or speaking, anxiety, among others.

Another problem with EA is that it's hard to count errors if there are multiple errors occurring close to each other. For example, “The man is drunken” for “The man has drowned” could be counted as one mistake (“is drunken” for “has drowned”), two mistakes (“is” for “has” and “drunken” for “drowned”), or three mistakes if we add (“drunken” for “drunk”).

CA: The aim of Contrastive Analysis approach is to study both L1 and L2 languages to collect data about differences and similarities between them to come up with those structures which students may find difficult to produce.

• The focus is on the language.

• This approach is linguistic-driven.

• Linguists study errors before they have occured.

EA: The aim of Error Analysis approach is to study the learner's perfomance when using the language, analyzing the errors made to see in which learning stage the student is in.

• The focus is on the learner's perfomance.

• This approach is data-driven.

• Linguists study errors once they have occurred.

While CA aims at preventing errors from happening, EA analyses errors to understand how learning is happening.

• CA errors attributed to negative transfer are not always produced by foreign language learners, whilst EA is based on the description and analysis of real errors made by speakers. 

• EA is motivated by Second Language Acquisition and focuses on the students’ learning process, unlike CA, which focuses more on teaching than learning.

• EA’s perspective is influenced by Mentalism and perceives students as an active and creative participant in their learning process. On the other hand, CA is influenced by the behaviorist assumption that a language is acquired by habit formation and the student is a “passive recipient of language stimuli”.

• From the EA’s point of view, errors are an important and necessary part of the learning process, as they are seen as the source to get a deeper understanding of it. Additionally, it is seen as evidence that the leaner is exploring the language on their own, since without errors, learners cannot improve their knowledge of the language. On the contrary, CA understand errors as an “interference” to learn L2. They are bad habits that must be eradicated.

Main objections presented against EA:

• It is difficult to classify errors, as knowing which ones are systemic errors (done because of L1 influence) or mistakes (errors of performance related to the circumstances of the moment of speaking) is ambiguous. During the performance, both correct and incorrect forms of L2 can happen at the same time. Additionally, assigning a psycholinguistic cause to an error is speculative, as it is not easy to determine whether an error is caused by language transfer or is affected by more contextual aspects. 

• It does not provide positive information about what the learner has actually acquired at that moment of speaking, it only focuses on negative evidence. To understand the learner transitional competence, it would be better to try to describe their language as a whole.

• It is possible that there are learners that do not produce certain errors only because they avoid using a specific set of difficult structures.

• Regarding error evaluation, the gravity ranking differs a lot among judges, as they examine errors with different sets of criteria

Secondly, Error Analysis does not predict errors, but it works with actual errors that learners have made in their L2. In comparison to the CA approach, the EAA states that learners errors are an important way of improving the learning process

Adriana Delgado

On the contrary, EA has as a main objective the description and analysis of actual learner errors in the L2. Another characteristic of this theory is that errors are seen as fundamental in order to acquire a language. However, EA is said to be problematic because there is no clear definition as regards what constitutes an error. This means that it’s difficult to spot which of them stem from a lack of knowledge of the language, and which are caused by other factors, such as situation, interlocutor, or the distinction between speech and writing.

There are two main objections to the Error Analysis approach. Firstly, sometimes, learners might produce errors that have nothing to do with their knowledge of the target language. These errors might be wrongly analysed by the EA. For instance, if an advanced English student is giving an oral presentation and, because of their nervousness, mistakenly conjugates a verb in the third person when talking about themselves, the EA could declare that the student has problems with verb agreement when it was only a mistake in their performance. Secondly, if the learner avoids using complicated structures because they don’t want to take risks by using complicated structures, the EA cannot describe if the learner knows about the topic or not.