maket_N_verjin.qxp


Meaning and Association

Approaching the problem of meaning byconsidering the responses to a word is a
psychological tradition. The history of psychology is
abundant in theories which have attempted to account for
“the mind” by appealing to doctrines of association: Locke
spoke of associations between ideas, Titchener - of
associations between sensations, Thorndike and Hull - of
associations between stimuli and responses. In all the cases
the associations referred to are theoretical, that is, they are
not observable or directly measurable in any way. They are
events of the mind or of the central nervous system,
depending on which theory is being considered.

Associationists have been concerned with the relative innateness of psychological
processes and the contents of the mind. They typically side with those who feel that most
of these processes and contents are learned. (Boring 1950:29) On the opposite side are
those who speak mainly of innate mental organizers, ideas, and the like, and who also
tend to be non-associationistic in their thinking. 

Experimental psychologists from the time of Wundt and Ebbinhaus have
considered meaning to be the function of associations. One thing could have meaning
only if it was associated with something else; the nature of meaning depended on the
particular associations and the multiplicity of associations involved. It should be noted
that these early theorists used the term “association” in a more or less technical sense. In
other words, associations were interpreted as important links between ideas or between
sensations. The desire for more precise experimental investigations of psychological
problems led to the idea of measuring the number and the kinds of associations. The
relationship between associations and learning is by no means a universally accepted
conclusion. It should be realized that something’s having association with another thing
is not identical with its having meaning in a non-linguistic sense. 

In psychology and psychiatry associations are used for the study of human
behaviour and mentality whereas psycholinguists are interested in the study of  verbal
associations to understand the peculiarities of mentality and its verbal expression.*

There are two basic ways of measuring associations to words. The first method
expounded by Kent and Rosanoff is regarded as the method studying the situation when the
subject is presented with a stimulus word and asked to respond with the first word he thinks
of. (Terwillinger 1968:166) When talking about the nature of associations, we need to refer
to associations given by a large number of subjects to a large list of stimuli. The degree of
commonness of the most typical responses varies from word to word but in some cases may
exceed 80 per cent. In other words, it is not unusual for a vast majority of the subjects to
agree upon the association given. Words vary in their distributions of associations - some
words reliably have associations more equal in their frequency of occurrence than others. 

59

Linguistics Armenian Folia Anglistika

Rouzanna Arakelyan



It is much more difficult to be precise about the qualitative nature of the
associations given. We find that synonyms (cross-angry, glad-happy, merry-happy,
boring-dull, attractive-pretty, astonished-puzzled, beautiful-pretty) and antonyms of the
stimuli (good-bad, interesting-dull, happy-unhappy, handsome-ugly, sad-happy) are the
most common associations. To some extent it is possible to divide associations into two
categories: those which reflect or respond to the referential characteristics of the word,
and those that reflect its uses in utterances. It is worth mentioning that the first type of
associations is more common in children rather than adults.

Very often subjects reflect certain mood or evaluative judgements which are
beyond the understanding of experimental psycholinguists partly because they are not
associations which are relevant to the common social uses of language. Among such
are the cases when the subject responds with a proper name to a stimulus word
(surprised-Newton). It should be noted that there are subjects who clearly ignore the
instructions and give associations which represent non-uses of the word in order to
confuse the experimenter. What they fail to realize is that by doing so they are giving
data potentially as useful as that given by more co-operative subjects. When subjects,
consciously or unconsciously, give words that are not associates of the stimulus, they
define the meaning of the word as accurately as if they gave associates consciously.
This phenomenon occurs because the meaning of words can also be defined in
contexts in which  the word is not used. No verbal subject can remain independent of
verbal stimuli; he may obey or ignore instructions, but he responds to the stimuli
nevertheless. 

In the other association method the subject is given a stimulus word and asked to
give as many associations to it as he can within a given period of time. The results are
presented in terms of the average number of associations given by the average subject.
This method has obvious advantages as a simple description of meaningfulness or the
quantity of associations. It should also be mentioned that the method described has one
methodological drawback: one can’t be sure that the experimentee is always really
responding to the stimulus word. It is possible that the subject is giving first an
association to the stimulus word, and then an association to the first reaction. These types
of associations are usually based on phonetic similarities (calm-sum, dumb, lucky-
feeling, dealing). This is called chaining of associations which is of little importance to
the expert if he is not interested in the quantity of responses. 

In “free-association” method the stimuli are removed from any normal context in
which they might ordinarily be used. But they are not free of contextual effects.
According to Kent-Rosanoff’s approach to associations as indicators of personality
variables, they are determined by  idiosyncratic contexts  chosen by the participants of
the experiments. 

The association methods while equally permissive with respect to context, permit
the subject to “use” the word, or respond to it, in any way he chooses – providing, of
course, that the response is another single word. The responses obtained, then, could be
more relevant to the actual use of the word and it  would seem that a subject might tend
to use the word in a way in which he was accustomed to using it, providing that the

60

Armenian Folia Anglistika Linguistics



instructions for the associations obtained in the so-called free-association studies may
well reflect or be related to the actual use of the stimuli. 

It is also clear that verbal associations do not represent the totality of what we
would call the meaning of a word. These associations are indeed uses of the word, but
they are uses made in the restricted environment of the psychology laboratory. It would
be right to say that, in general, the associations obtained from the subject in the
association method represent a set of verbal descriptions of how the subject uses the
words. The associatons are clearly related to the meaning of the words. Thus, the
assumption that a word which is used in different ways will call out different verbal
descriptions of those uses. Accordingly, it is true that the number of different verbal
descriptions will be proportional to the number of different uses. (Popova 2006:12)

There are two different situations in which we must evaluate the above mentioned
assumption. First, let us consider the  case in which the subject gives many responses to
a single stimulus. Here it seems quite safe to assume that the more meanings a word has
for the subject, the more verbal description he will give of it. Presumably we might infer
something about the number of meanings from the number of responses given. 

When each subject gives one response per stimulus, the number of different
descriptions is determined across subjects, rather than within one subject. If we assume
that an individual has several possible uses for a word and he must pick one of them to
use in an otherwise ambiguous situation, there are certain things we expect to find. If we
presume that some uses are more common than others, we might assume that he would
pick accordingly from among the possible varieties. In a large number of subjects, the
probability distribution would be identical to the individual. So if language  is to function
as language, it must be shared by members of the linguistic society. Since social
agreement is by definition a characteristic of language, we might expect to find
agreement among associations as well. (Zif 1960:68)

We assume that the number of different associations given by either association
method is proportional to the amount of the stimulus word. Likewise, the obtained
probability of any given association is proportional to the probability of the particular
word use that the association in question describes. 

It becomes clear that verbal associations do not directly constitute meaning; they
are in reality descriptions of meaning or use. We propose to be neutral about how much
of the mind is learned and postulate the following idea: whether it is learned or innate,
it is properly describable in associationistic language. 

Notes:

* One of Ch.Darwin’s cousins, sir F.Galton, was the first to conduct a
psycholinguistic experiment. He chose 75 words, wrote them down on separate
cards and did not touch them for a few days. Then he took the cards and looked at
them for some time trying to define how much time passed since his seeing the
word and the first reactions coming to his mind. He wrote down the reactions, but

61

Linguistics Armenian Folia Anglistika



never published them, because he thought that it would uncover the essence of
man’s mind so vividly and truthfully that it would become world’s property. Since
then this method has been widely used by psychologists and psychiatrists, and later
on by psycholinguists for the purpose of studying personality’s psychics and
mentality. 

References:

1. Boring, E.G. (1950) A History of Experimental Psychology. New York: Academic
Press.

2. Popova, T. (2006) Assostiatsivniy Eksperiment v Psikhologii. Moscow: Prospect.
3. Terwillinger, R. (1968) Meaning and Mind: A Study in the Psychology of Language.

New-York-London: Oxford University Press.
4. Zif, P. (1960) Semantic Analysis. Hove: Cornell University Press.

Ü߳ݳÏáõÃÛ³Ý ¨ ½áõ·áñ¹Ù³Ý ÷áËѳñ³µ»ñáõÃÛáõÝÁ

¼áõ·áñ¹áõÙÝ»ñÇ áõëáõÙݳëÇñáõÃÛáõÝÁ Ñ᷻ɻ½í³µ³Ý³Ï³Ý ¹Çï³ÝÏÛáõ-
ÝÇó Ýå³ëïáõÙ ¿ Ùï³ÍáÕáõÃÛ³Ý ¨ ¹ñ³ ³ñï³Ñ³ÛïÙ³Ý Ó¨»ñÇ ³é³í»É ѳ-
Ù³å³ñ÷³Ï Ýϳñ³·ñáõÃÛ³Ý Ý»ñϳ۳óÙ³ÝÁ: ê³ Ñݳñ³íáñ ¿ Ñ᷻ɻ½í³-
µ³Ý³Ï³Ý ·Çï³÷áñÓÇ ÁÝóóùáõÙ Ù³ëݳÏÇóÝ»ñÇ ÏáÕÙÇó Ý»ñϳ۳óíáÕ
µ³é³ÛÇÝ ½áõ·áñ¹áõÙÝ»ñÇ ¹³ë³Ï³ñ·Ù³Ý áõ í»ñÉáõÍáõÃÛ³Ý ÙÇçáóáí: ²ñ¹-
ÛáõÝùáõÙ å³ñ½ ¿ ¹³éÝáõÙ, áñ µ³é³½áõ·áñ¹áõÙÝ»ñÇ µÝáõÛÃÝ ³ÝÙÇç³Ï³Ýá-
ñ»Ý ϳåí³Í ¿ µ³é»ñÇ Ûáõñ³óÙ³Ý ¨ ¹ñ³Ýó ÏÇñ³éÙ³Ý ³Ýѳï³Ï³Ý µÝáõÛ-
Ãáí: ì»ñçÇÝë ¿É áñáßíáõÙ ¿ û° É»½í³Ï³Ý ¨ û° ³ñï³É»½í³Ï³Ý ·áñÍáÝÝ»-
ñáí: ¼áõ·áñ¹áõÙÝ»ñÝ ³ÛÝ Ï³ñ¨áñ Ùï³Í³Ï³Ý ϳå»ñÝ »Ý, áñáÝù ³å³Ñá-
íáõÙ »Ý É»½í³·Çï³ÏóáõÃÛ³Ý Ï³ñ·³ÛݳóáõÙÁ:

62

Armenian Folia Anglistika Linguistics