In a multimodal expression signs are used to replace a word in a sentence and therefore “contribute semantically to the expression” (Green & Wilkins 2014, p.238). For example, the Arandic spoken language has a semi-equivalent of the HIT sign, “atwem,” while sign language can offer three different modifications of the HIT sign to convey different meanings. Thus, by orienting the sign in space in three different ways, a person could show “cutting wild honey from a tree,” shape “an artifact such as a wooden plate,” or represent “fighting” (Green 2014a, p. 140). Relying on both signs and speech allows the speaker to “achieve an appropriate economy of expression or particular effect on the addressee” (Kedron 1987, p.35). Green and Wilkins noted that it is possible to replace, for example, a word for not at all/no “arrangkwe” with a manual sign that has the same meaning. Therefore, if it is necessary to convey the proposition that there is no water, the phrase kwatye-ke arrangkwe (water-DAT nothing/no) 'there is no water', can be transformed into kwatye-ke followed by a sign for " nothing/no/negation” (Green & Wilkins 2014). Such a combination of two modalities saves the speaker time and effort. However, it is not clear what particular effect this multimodal utterance has on