This paper sets out to explore the expression of vagueness in eighteenth and nineteenth century scientific writing, specifically in the Corpus of English Life Sciences Texts. Following Zhang’s (2015) pragmatic-oriented approach to vague language, vague expressions are viewed through the lenses of elasticity. This notion applies to the strategic use of vagueness insofar as it provides a space for the negotiation of pragmatic meanings in writer-reader interaction. The exploration of elastic language involves four lexical categories: (i) approximate stretchers, i.e., approximators and vague quantifiers, (ii) general stretchers, i.e., general terms, placeholders and vague category markers, (iii) scalar stretchers, i.e., intensifiers and softeners, and (iv) epistemic stretchers. The Coruña Corpus Tool has been used to elicit data, but classification and contextualised interpretations have necessarily relied on manual analysis. Findings reveal that elastic language fulfils a variety of relational functions in scientific writing, including making generalisations when the information is either not available or relevant for the purposes of communication, marking shared knowledge and group membership, or self-protection.