A fist is a noun. To persist is a verb. Yet a clenched fist can last only a moment, while persistence can continue for years. Benjamin Lee Whorf used examples of this kind in “Science and Linguistics,” published in 1940, to challenge the idea that English divides experience neatly into enduring things and temporary actions. The grammatical classification and the duration of the event do not line up. His point was that familiar categories can seem natural precisely because their users rarely inspect them. Whorf, “Science and Linguistics”.
That is an arresting observation. It also leaves a further question: what happens in the mind when people use different categories? Noticing that languages organize descriptions differently starts an investigation. It does not finish one.
The linguistic record comes first
Edward Sapir's 1929 “The Status of Linguistics as a Science” located language study among anthropology, history, sociology, psychology, and philosophy. Its abstract argued both for the importance of linguistic methods to those fields and for linguists' attention to wider human conduct. Language was worth studying as social behavior, with consequences beyond a catalog of forms. Sapir's original article and abstract.
The attraction of comparing languages is easy to understand. Your own ordinary forms rarely feel like choices until another system makes a different choice. Consider a hypothetical editor changing “The committee rejected the proposal” to “The proposal was rejected.” Both are grammatical English. One puts the committee first; the other moves the proposal into that position and leaves the actor unspecified. Neither sentence licenses the conclusion that its writer cannot understand committees or agency.
The comparison nevertheless creates a precise problem. Would readers allocate attention differently? Would they later remember the actor differently? Would a difference disappear when the actor was already important to the story? A linguistic contrast identifies candidate information to investigate; the experimental question concerns what people do with it.
From a striking example to an outcome
The early argument's enduring challenge is methodological. If words participate in thought, asking someone to describe their thought may reproduce the very linguistic patterns under investigation. A researcher who hears a different description still has to decide whether the person experienced the event differently, remembered it differently, or simply described it in a conventional way.
Suppose, in a hypothetical study, two groups use different everyday categories for the same set of objects. Ask them to name the objects and the naming difference is almost guaranteed: that is how the groups were selected. Ask them to learn a new grouping, detect a changed object, or recognize an old one after a delay, and the experiment begins to address a further question.
Even then, the outcome matters. Accuracy is the proportion of correct responses. Response time is how long the response takes. A physiological measure records a feature of bodily activity, which needs its own interpretation. These measurements can constrain an explanation; none comes with a built-in translation into “worldview.”
A broad historical proposal becomes more informative as its components become separable. The linguistic description specifies the available distinction. The task supplies a situation in which that distinction might help. The outcome tells us whether the predicted difference appeared.
A color difference without a color instruction
A 2009 study by Guillaume Thierry and colleagues compared native Greek and English speakers using an electroencephalographic measure of responses to visual changes. Greek speakers conventionally named the light and dark blue stimuli with different terms; English speakers used blue for both. Participants watched a stream of circles and responded to occasional squares. They were not asked to judge the color contrast.
The researchers reported a larger visual mismatch-negativity response to blue than green changes in the Greek group; the English group did not show that same blue–green difference. They interpreted this as evidence for an influence of linguistic distinctions on relatively automatic color processing. Thierry and colleagues, “Unconscious Effects of Language-Specific Terminology on Preattentive Color Perception”.
The experiment is interesting because its instruction and its measure pull apart. The participant's job concerned shape. The researchers' question concerned responses to color changes. That design addresses an objection to tasks in which people might deliberately use color names to select an answer.
It still compares groups with different language histories. Language, education, cultural experience, and other background factors were not randomly assigned. The recorded response also does not tell us everything about the appearance of blue in conscious experience. The result supports a particular argument under a particular design; describing it accurately preserves its interest.
What a stronger account has to explain
An explanation should account for both an observed effect and its boundaries. If a proposed mechanism is attention, changing what a participant attends to should matter. If it is memory, a task with all the relevant objects continuously visible may behave differently from a task requiring retention. If it is active use of verbal categories, occupying verbal resources offers one way to probe it.
These are methodological inferences, not a report that every such manipulation always produces the same result. They show how an account can become vulnerable to evidence. A mechanism that explains any possible outcome cannot help distinguish competing possibilities.
This also changes how to read the history. Whorf's provocation about nouns and verbs directs attention to unnoticed classification. Experimental work asks a different, narrower question about performance or processing. The later question owes something to the earlier one without turning an early essay into the results section of a modern study.
The fist and the persistent action remain a good puzzle. What they give us is a reason to examine our categories. What an experiment adds is a way to find out which consequences those categories actually have.
Loading comments…