Abstract
Prescriptive grammarians advise against the use of the quantifier many in informal positive sentences, unless it is modified by too, so, or as. However, because these grammarians may not have conducted a thorough corpus-based analysis, such advice may be unsound. This is why this article attempts to identify the actual constraints on the use of many, by searching corpora for data for many and its competitor a lot of, with plural count nouns in positive sentences. Conducted within the conceptual framework of cognitive linguistics, the analysis suggests that quantities denoted by many are construed as heterogeneous and discrete, hence the relative affinity of many for nouns of place and time, as part of adverbial phrases. This core meaning may also account for many's relative affinity for personal nouns in subject position. Unlike a lot of, many seems to associate awkwardly with homogeneous substances, which may be why it is rarely found in object noun phrases.