View Our Catalog
and Focus in American Sign Language|
Fortunately, the importance of learning vocabulary in context and of being able to use the appropriate constructions is generally understood. We know that replacing English words with “equivalent” Spanish words does not yield an effective translation, as evidenced by the hilarity (or downright weirdness) that can ensue when performing such “word-to-word translations” via web-based searches such as Google Translate. It goes without saying that students of a language who learn vocabulary but have no clear understanding of the ways in which constructions at the phrasal, sentential, and discourse levels interact with that vocabulary will be ineffective at conveying their true intent.
Unfortunately, this seemingly commonsense understanding that meaning in language stems from so much more than vocabulary alone is somehow less self-evident when the languages under consideration are a spoken language and a signed language. Formal linguistic analysis of signed languages is a relatively recent undertaking (generally considered as getting its real start with the work of William Stokoe in the early 1960s; Stokoe 1960; Stokoe, Casterline, and Croneberg 1965) and as such the extent of what we know about grammatical and discourse structures remains quite limited. Though our foundation is strong and the field is blossoming as it reaches the half-century mark, linguistic knowledge about American Sign Language (ASL) as used by those who acquire it natively from their parents has not been able to keep pace with the demand for teaching ASL to aspiring students. This, in combination with myriad sociological and biological facts about deaf lives and signed languages (Padden and Humphries 1988; Van Cleve and Crouch 1989; Lane, Hoffmeister, and Bahan 1996; Padden and Humphries 2005), conspires to create a situation in which vocabulary teaching takes precedence. As a result, a true grasp of the meaning conveyed at other levels is often missed.
passive voice as a case study
Take passive voice as just one example. Many materials designed for teaching ASL and/or for teaching ASL/English interpretation and transliteration (Kelly 2001; Mikos, Smith, and Lentz 2001) maintain that ASL does not have passive voice and that agents must therefore be in focus. Students are taught that agents should always be overtly specified in prominent positions (i.e., at the beginning of an utterance and expressed as either a topic or the subject), and interpreters are encouraged to produce “active” signed translations of spoken utterances expressed in passive voice.
If beginning users of ASL are going to be interpreting, there is good reason for teaching them to restructure English passives into active ASL structures. Literal translations of passive utterances in English generally leave out the morphemes that mark verbs as passive (a form of to be and the past participle of the main verb), resulting in an expression that literally reverses the roles of agent and patient.
Consider, for example, this short text about an experiment on REM sleep:
REM Sleep Experiment
The main clause of the second sentence is “volunteers were taught.” A literal translation that followed the English word order and omitted the passive markers would potentially be volunteers teach, incorrectly stating that the volunteers were the ones doing the teaching rather than the ones learning the new skill.
The following sentence in the English text (“some of them were awakened”) is also in passive voice, and a literal translation omitting the passive marking would again result in a skewed meaning, though in this case the effect on the meaning is quite different. Here an attempt at translation that ignores the content conveyed by the passive structure would result in the ASL utterance some pro.pl→volunteers wake-up. Because the volunteers are expressed as the subject of the active verb wake-up, the meaning conveyed is that the volunteers woke up of their own accord, which is not the meaning conveyed in the original English sentence and in fact skews the entire understanding of the experiment.
Thus, teaching beginning students of ASL, who are producing word-by-word literal translations of the English sentences in their heads, to recognize agents and express them actively in ASL makes sense. However, the underlying assumption that in ASL agents must always be in focus is untrue.
If it were true that agents must always be in focus in ASL, one would expect ASL users, when asked to translate English passive sentences into ASL, to produce only utterances with an agent-focused construal. They might express all assumed agents overtly as subjects. For English prompt sentences with by-phrases, one would predict that ASL users would reorder the entities to produce an utterance with the agent in subject position.
However, in data collected for my dissertation, when native ASL users were asked to do precisely that — translate English passive sentences into ASL — only 11 percent of the utterances they produced had the agent in focus. Within the set of twenty sentences containing passive-voice clauses, three of the passive prompt sentences even included overt mention of the agent (i.e., in a by-phrase). Even from these prompts, where rearrangement to produce an ASL utterance with the agent in subject position is remarkably straightforward and would be expected, participants reordered the entities only half of the time. Even more striking, out of eighty given opportunities, signers chose to overtly express an assumed agent in only four instances.
Clearly, ASL users do have ways to express a construal in which the agent is not the primary focus. The goal of my dissertation research and this book describing those results was to determine the linguistic strategies that ASL users employed to express a meaning similar to that expressed when English speakers use passive voice. The data showed that, in ASL, just as in English and other languages, various linguistic factors influence the level of focus with which entities are construed. In addition, ASL has multiple mechanisms for achieving agent impersonalization, and each mechanism, with its unique form, results in a slightly different construal.