The concept of eonmunilchi (the unification of the speech and writing) reminds us of the unification of written and spoken language. However, from the late 19th century to the early 20th century, two problems were intervened in the argument of eonmunilchi. One is about the use of certain characters, and the other is regarding the styles that are used to complete the sentences. But this paper treats the problem of styles of the sentence level, not the problem of characters. In Korea, it is often said that the typical example of unification of written and spoken language at the level of sentences is the replacement of '-deora (-더라)' with '-eotta (-었다),' However, in modern Korean, '-deora' is mostly used in the spoken language whereas '-eotta' is rarely used in the spoken language. If '-eotta,' a typical example of ending expression of eonmunilchi is not related to the spoken language, then what is the fundamental meaning and effect of eonmunilchi? The basic effect of eonmunilchi occurs when it is difficult to suppose the relationship of speaker/writer and hearer/reader or even the context itself where a locution takes place. In other words, erasing the trace of speaker or writer in the text so that the sentence itself could carry out its own meaning is regarded as the effect of eonmunilchi.