Common use of Methods and Materials Clause in Contracts

Methods and Materials. The first set of studies was designed to better docu- ment and quantify the variations between British and American English in the use of plural number when an agreement target’s controller or antecedent has a collective head. We gathered two types of data from British and American speakers and writers. First, to assess verb agreement using matched collective and noncollective materials under comparable conditions, we elicited spoken sentence completions from American and British college students. The collective nouns sampled were chosen from a diction- ary of collectives (▇▇▇▇▇▇▇ 1985) to represent a range of human (team, committee), nonhuman animate (herd, flock), corporate (government, association), and ostensibly inanimate (fleet, forest) groupings. Second, for a subset of these collectives we carried out counts of singular andplural verb andpronoun agreement with collective controllers in the Wall Street Journal corpus and the British National Corpus. SPOKEN SENTENCE COMPLETIONS. To compare the incidence of plural agreement after collective controllers with the incidence of plural agreement after semantically related plural and singular noncollectives, thirty-nine students at Michigan State University and thirty-nine students and research workers at Cambridge University provided spoken sentence completions. The completion test was assembled from ninety-six triplets of semantically relatednouns (see Appendix A). Each triplet consistedof a collective (e.g. army), a semantically related noncollective singular (e.g. soldier), and the correspond- ing plural (e.g. soldiers). Three lists of ninety-six simple definite noun phrases were assembledfrom these triplets, with one noun in each noun phrase (e.g. the army). Every list contained one noun from every set and an equal number of nouns of each of the three types (collective, singular, andplural). Across the three lists, every noun occurred just once. The order of the nouns within lists was random, constrained so that there were no more than two successive occurrences of the same kind of noun. The same random order was used for all three lists, so that nouns from the same triplet occurred in the same ordinal position in every list. Each list began with the same four practice items, consisting of two noncollective singulars and two noncollective plurals that differed from the noun phrases used within the lists. The phrases were presented to participants individually under computer control, each phrase appearing centeredon the computer monitor. Where there were spelling discrep- ancies in American and British (neighbor/neighbour), the spelling presented was the appropriate one for the dialect. Participants were asked to read the phrase aloud and complete it as a simple sentence, as fast as possible with the first thing that came to mind. The speakers were instructed by example to use completions consisting of the copula BE and a predicate adjective. On each trial, when the participant began to talk, the experimenter clearedthe computer screen. At the completion of the trial, the participant pressed the computer’s mouse to move on to the next phrase. An additional sample of spoken completions was gathered from thirteen British students enrolled at Michigan State University, identified as British nationals by the campus administration. Their durations of residence in the United States varied. Each of the British students received the entire set of 288 definite noun phrases arranged in one of six random orders, preceded by the same four practice trials. In other respects the procedure was the same as described above. The participants’ responses were recorded, transcribed, and scored. The scoring noted whether the verb usedwas singular, plural, or other. The ‘other’ category coveredcases in which the verb couldnot be unambiguously scored as singular or plural or the subject noun phrase was inaccurately produced. CORPUS COUNTS. The part-of-speech tagged Wall Street Journal corpus and British National Corpus were searched for occurrences of the subset of collective nouns listed in Appendix B. To better equate the subject matter of the American and British texts, the search in the British National Corpus was restricted to the domain of finance and commerce. When a collective served as the subject of a clause or as the same-sentence antecedent of a third person pronoun, and was not part of a proper name, the verb or pronoun agreement target was hand-coded as singular, plural, or unspecified (for verbs with morphologically invariant number, such as past tense verbs). The search in the Wall Street Journal corpus was exhaustive. In the British National Corpus, the number of tokens for each collective was set at a maximum of 300, sampled at random from all of the texts within the domain. For the two sources, Appendix B gives the distribution across the collective nouns of the incidence of verb and pronoun tokens with unambigu- ous number.