Looking back to human language history, it is almost always the case that spoken language precedes written language. Also, developmentally, children learn to speak much earlier than they learn to write. Hence, speech is usually considered to be the primary linguistic modality, whilewritten language is taken to be a cultural artifact used to document the oral speech. This is an orthoxy in the history of Linguistics. Saussure (1916) contends that the only reason for the existence of writing is to represent speech. Similarly, Bloomfield (1933) claims that “writing is not language, but merely a way of recording speech by visible marks” (p. 21). Contending that language is an aspect of human biology rather than a cultural aspect, Bloom and Pinker (1990, p. 707) stressed the contrast of the two modes of communication, i.e. speech and writing: “language is not like writing or the wheel”. Bromberger (2012) explicitly claims that “written words are cognitively parasitical on spoken ones...and thus though spoken words can be acquired through exposure, written ones require prior ability to speak” (p. 9). Theorists in general think explanations for the facts about linguistic structures in spoken language can be generalized to other modalities. Many linguists have utilized the features of spoken language observed to draw inferences about the nature of the human language faculty. For example, Chomsky thinks that language is a function of mapping the phonological representation to the semantic representation. The principles that govern spoken language also govern linguistic phenomena in other modalities. In this talk, I will argue that the traditional picture of taking spoken language as the primary modality to understand other modalities of language is mistaken. I propose to look at two modalities of both sign language (written and signed) and logographic languages like Chinese (written and spoken). By showing that there are cases in which meanings cannot be disambiguated only by analyzing the spoken sounds, but rather we need to look at the written words, I argue that there might be different procedures of mapping forms of different modalities to meanings. I will also present arguments for why such mechanisms are distinct from the bilingual case in which onemechanismmaps two different linguistic forms to the same meaning.