It is very important diagnostically to understand psychological process what errors are appeared as well as analysis of phenomena of them such as erasion, morphological errors, collocation errors words generated by foreign learners of Korean. Metaphorically speaking, it is alike that the doctor should make a diagnosis on why patients hurt rather than know they are sick. This paper clarifies the results for this view. I firstly argued that the errors should be considered that they do not belong to either L1 or L2. because they as themselves have the characteristics of the intermediate language. Therefore, I assert to use the terms of errors of L1 approaching interference, ones of L2 approaching interference, and L1 and L2 approaching interference instead of using interlingual errors, intralingual ones and complex ones. The second, this paper clarified that errors of semantic similarity, errors of idiomatic collocation can be vested in error type of L1 approaching interference, not errors of L2 approaching interference according the difference of languages. It is imply that causes of error each can be possible to vary, even though superficially can be vested in the same type. The third, there is the type of error that L1 rule incorrectly transfers L2 rule, that means, the type of `correspondence of L1 rule to L2 rule`, even though an existing discussion didn`t mention it. Fore example, it is Errors of Sino-Korean numeral collocation generated by Japanese learners. The fourth, Errors of omission or incompletion and errors of word coinage are resulted from wrong correspondence of L1 lexicon to L2 syntactic component, not the correspondence of L1 lexicon to L2 lexicon. The fifth, the phenomenon of redundancy errors is not the type of complex interference corresponding L1 lexicon to L2 lexicon, but ones corresponding different component to each especially Argentan learner. It also means that each one in which superficially has the same phenomena can be nested on the other type.