A fairly common phenomenon of inflection is morphological haplology, in which an affix or clitic is absent when the adjacent part of the stem is homophonous to it, e.g. Eng. poss. pl. boys', not boys's. Data are adduced from numerous affixes and clitics in eleven languages. Haplology may be a major or minor rule; it may show lexical diffusion; and it is subject to various specifications on how similar the ‘homophonous’ part of the stem must be, including degrees of phonological similarity and whether it too is a morpheme. These data are traditionally handled with a rule of deletion or haplology. Such an analysis is undesirable because of the false step required—the fact that deletion does no more than recover the input form; it does not explain why it is always the affix that is deleted, rather than part of the stem. It also does not explain why a haplologized clitic must always dominate the other morpheme/phonological sequence. The alternative is not to add the affix/clitic at all. This is best viewed as a case of vacuous rule application, where the rule may take the input as already satisfying the conditions of the output. Such a proposal can be made only within a morphology that stresses the prosodic nature of morphemes, as in autosegmental morphology. The origins of this phenomenon in child language and in language change are discussed.

Relational Grammar and Invariant Ordering Grammar both take the linear order of constituents to be a superficial property of sentences which can be predicted by a set of so-called word-order or linearization rules. The constraints which these two theories have imposed on their respective linearization rules are examined, and are shown to make wrong predictions about the possible word orders that active and passive sentences exhibit across languages. Furthermore, it is shown that these constraints cannot be amended in any obvious way to give the desired results. To account for the facts in question, a new mechanism is introduced and briefly argued for.

The syntactic rule of Restructuring explains the syntactic properties of so-called ‘Clitic Climbing’ sentences in Italian. However, Restructuring alone fails to account for the semantic properties of such sentences. This paper argues that lexical governance of Restructuring is inadequate. Instead, a syntactic rule operating freely whenever its structural description is met, and followed by semantic interpretation rules, not only adequately but insightfully handles the data. A crucial claim is that the derived structure of the verbal complex in a restructured sentence is identical to that of a complex of auxiliary + past participle, so that the same semantic interpretation rules apply to both.

Friedrich & Redfield (in Lg. 54.263–88, 1978) have argued that Homer's heroes had their own idiolects. This view is untenable: it is inconsistent with the practices of oral poetry and with the classical doctrine of êthos. Their stylometric tests are unconvincing because of inadequate sampling, an unreliable textual tradition, and a defective rationale in applying stylistic criteria.