Kiparsky (1982) proposes two different classes of instrumental denominal verbs in English: the hammer-type (1) and the tape-type (2). These are distinguished by whether an adjunct PP can introduce a distinct instrument argument, different from that named by the verb, to the clause:

(1) Lola hammered the metal / hammered the metal with her shoes.
(2) Lola taped pictures to the wall / *taped pictures to the wall with pushpins.

Kiparsky’s analysis of these purported classes is that tape-type verbs derive from nouns in the lexicon, with resulting meanings based on the meaning of the underlying nouns. Hammer-type verbs are not derived from underlying nouns so their meanings are not tied to specific noun roots. Arad (2003) integrates this idea into a non-lexicalist analysis, wherein hammer-type roots become verbs by merging directly with v (hence denoting actions which need not involve actual hammers), whereas tapetype roots acquire a nominal interpretation by merging with the functional head n prior to merging with v (hence denoting actions requiring actual tape). We suggest that no account of this distinction is necessary, as the distinction is spurious. Verbs of the tape-type do not necessarily entail use of the conflated root:

(3) Lola taped the poster to the wall with band-aids / mailing-labels.