Answers are grounded in this paper and cite their sources.
The typical way in which a mass noun is made countable in English is by putting a count noun before it and connecting the two by of (e.g. a piece of information ). When these binominal NPs are modified, the adjective has the choice of going on the first noun (e.g. an additional piece of information ) or on the second (e.g. a piece of additional information ). As the count noun has a largely quantifying function and as the mass noun is the semantic head, adjectives may be expected to select the mass noun (N2 modification) rather than the count noun (N1 modification). Contrary to expectation, a COCA analysis reveals that N1 modification by far outnumbers N2 modification. This Anticipation Effect is brought about by a general processing asymmetry whereby later elements are more strongly inclined to occur early than earlier elements are to occur late. This processing principle operates within a specific linguistic context with a relatively weak syntactic boundary between the first noun and the PP and a deeper embedding of the adjective in N2 than in N1 modification. Since depth of embedding is an index of availability, the first noun is more available for modification than the second.
No takes yet. Share an insight, caveat, or question.
Thomas Berg (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: