May 15, 1996

Linguistic Power

Language use is one of the most amazing talents which humans posses. It is a complex process involving everything from phoneme recognition to the organization of ideas into meaningful discourse. In this paper I will discuss some of the important issues surrounding the comprehension and production of language and the learning of language. Beginning with the basics of what it means to know a language and moving on to the processes involved in its production and comprehension, I will attempt to summarize some of the findings in the field of linguistics. Due to the immense size of this field, I will in no way be able to credit all those who deserve it for their contributions. I will instead point out any contributions and their sources that I am aware of, and include some other works which I referenced while forming this summary.

At the most basic level, language relies on the physiological ability to detect and hear verbal communication (putting aside, for the moment, languages such as American Sign) and the physiological ability to produce the correct sounds. These ideas will be referred to as the auditory and articulatory abilities respectively. It has long been known that a baby will make every possible sound unit, or phoneme, while babbling during the first year, or so, of life. This full use of articulatory range occurs at a very young age and once the child begins to have some experience with the other people in its life it begins to learn which phonemes are being used around it and the production of these phonemes becomes reinforced. Soon the baby will cease to produce the phonemes which it does not hear around it. During this time, the baby is also learning how to hear the different phonemes and thus is exercising its auditory ability.

Within the first two years the child has learned how to produce most of the phonemes which are being used around it as well as to recognize them when heard. One clear way in which you can see that the child is still developing its auditory ability is that children often pronounce a word like ‘fish’ as ‘fiss’ and, when told that the word is ‘fish’ not ‘fiss’, the child will insist that is what they had said. One possible interpretation of this data is that the child has not yet mastered the auditory recognition of some of the phonetic sounds involved in the pronunciation of ‘fish.’ It is important to remember, however, that the auditory ability in reference to the speech of others does not necessarily translate to that of oneself. This is to say that a child may still have difficulty in recognizing its own speech even though it can detect the difference in ‘fish’ and ‘fiss’ when spoken by others. This example also shows that children may often take more time to learn how to put together various combinations of phonemes compared to that required to master the basic phonemes. This is, in part, due to complications in real speech such as coarticulation.

Coarticulation refers to the fact that when we produce multiple phonemes in succession various pieces of them become articulated at the same time. To demonstrate this point, you can take the sample of the speech of the word sue. If you study the spectrogram of the sample you will find that it clearly splits, near the middle, into two sections and you would expect the sound of the ‘s’ to be in the first part and the ‘oo’ to be in the second part. In fact, a listener who is presented with just one section will be able to identify each sound. The interesting part is that when presented with the ‘s’ section they will also be able to tell you the vowel which followed it because of the coarticulation of phonemes when spoken in succession, as they are in normal human speech.

We now have children who have mastered most combinations of phonemes (although some physiologically more difficult sounds may still present problems) and who can accurately identify the speech of others and themselves. They now have the basics underpinning language use. To know a language certainly includes these basic skills (again, putting aside cases such as sign language) although it must also contain a much more extensive amount of knowledge on meaning. The concept of relating any concrete item to its mental representation, or meaning, is called semantics and this is where the real ability of language comes into play. This will also help explain the use of sign language and other non-verbal communications.

The lowest stage in semantic understanding is that of lexical knowledge. The idea here is that we have a large storehouse of information, called the lexicon, which contains all of the items of our language which we know. Some theories describe the lexicon as being a collection of nodes which have various pieces of information attached to them. The auditory form of a word may be attached to it so that we can recognize it when heard. The articulatory form may be there so that we can express it. The semantic meaning of the word may be attached so that we actually understand what we are saying or what is being said to us. This gets at the heart of what it is to know a language. One of the most intuitive conceptions is that language is the collection of information centered around these lexical nodes.

Representing the lexicon in this way allows for a very general view of language which accommodates many non-verbal means of communication. In sign language, people must still learn the forms of words in order to recognize them and also in order to produce them. Those forms would merely involve physiological processes with the hands and arms, and not the vocal tract and such. Assumably, ‘speakers’ of sign language do still have similar lexical entries inside of their minds. One can also see, in cases of Aphasia in which the person is still able to write but not speak, that they have not lost the representations of recognition, only those of production which gives strength to this multifaceted view of lexical items.

An interesting item which has fascinated researchers involved in the study of the lexical system is the phenomena of priming effects. This refers to research involved in phonetic and semantic priming. An example of phonetic priming is that a subject who has been exposed to the word ‘car’ can more quickly recognize the word ‘card’ even though they do not have any semantic relation. In contrast, semantic priming involves the quicker response to ‘truck’ after the primer, ‘car’, because of semantic similarity even though they are phonetically quite different.

To say that there has been a lot of research on the various forms of Aphasia is to barely hint at the truth. It is a fascinating arena, but one which I will not have a chance to go very deeply into. When studying those afflicted with one form of Aphasia or another, the patient has usually suffered damage to some part of their brain. When evidence was first found that these damaged regions corresponded reasonably to the resulting Aphasia that they had led researchers to the idea, while seemingly valid on the surface, that the human brain was modularized for linguistic processing. That is to say that when we saw people with damage to the same area of the brain having the same, or similar, Aphasia we were seeing that particular module being broken! It has now been shown that modularity is unnecessary in explaining such disorders. Some work related to the competition model (MacWhinney, Bates 1989) has shown that similar deficiencies can be reproduced through global deficits in processing ability which is what the damage to the brain is believed to cause.

So clearly, as stated previously, the auditory and articulatory abilities are part of one’s knowledge of a language with respect to each lexical item in that language and some associated meaning. However, if this was the whole picture than we would have children mumbling words with no comprehensible relationships between them! The next stage in semantic understanding is syntax, which I will use to generally refer to the structure of words within an utterance to form a meaning. Some researchers, Chomsky being the most well known, have proposed the idea that there is some sort of universal structure to language which is apart from any one language we may learn and that it can be used to interpret any given language. Whether this is true or not, it is clear that there are some major commonalities amongst languages in terms of their structure.

The most important role of syntax is to allow two people to communicate information which is understood as it was intended. When a person says something to another person they are trying to express a thought, a particular collection of meanings which are being represented by items in their lexicon. If the listener does not share the same syntax then they will not be able to recreate the proper meaning. If the listener has different meanings associated with the lexical items then they will also not be able to recreate the proper meaning, however this is not something you can account for except by establishing standard dictionaries of meanings. If a person utters 2 nouns and an action such as “Dogs chase cats” then the listener must know that they mean the dogs are doing the chasing and not the cats. The information that tells the listener, and for that matter the speaker as well, the role of the words in an utterance is the syntax.

As it has been well documented in research on the competition model (MacWhinney, Bates 1989), English relies heavily on word order in its syntax to determine agency. Various languages rely on various cues, but all of them do have some fixed structure to these cues which can reliably describe an utterance’s meaning. Learning the particulars of a language’s syntax and its cue structure is one of the largest and most important steps in learning a language. Experiments have shown, as the competition model has predicted, that when learning those cues there is even a set order in which one will learn them, relying first on those that are most available and then later focusing on the ones which are most reliable, especially in conflict.

Taking all of this into account, we have several abilities: we can produce the sounds which are used in our language; we can produce them in combinations as they are used in our language; we can recognize the various phonemes and the words which they form; we can understand the meaning of the words and the meaning of the sentences in which those words are found; and we make use of the various cues in our language to help understand the meaning. These are what I would call the basic facts which are what we have come to know when we have come to know a language. These are the items that a native speaker of a language must come to know in order to be competent.

Above all of these is the ability to create utterances which have never been spoken before, or at least not to the knowledge of the speaker. This is one of the most historically ‘sacred’ facts about human language use though it seems unclear to me how this is not plainly intuitive with the current understanding of grammar rule formation. It has been found that you can use a small, finite set of rules to put together an infinite number of valid sentences. One way or another, this is an important feature of human language and one that allows us to continuously expand our knowledge by allowing new and creative utterances and therefore new and creative ideas. The question of whether this creativity and general linguistic ability is innate or not has been long debated and, as with Aphasia, we will not be able to cover it in any detail. We will continue with the summary with the understanding that, innate or not, we do eventually gain these abilities.

With the amount of diversity in our world it is quite clear that the knowledge of a second language greatly increases your ability to adapt to life. As such, the study of second language learning, bilingualism, and multiculturalism are of great interest to modern linguists. In order to learn a second language, and especially to become proficient enough in it to be considered a bilingual, you must overcome many mental hurdles. All of these aspects of language which we have discussed have been learned in a very extreme way. Put another way, language becomes the second-nature, easy process that it is through an amazing level of experience at using it. In order to learn a completely new set of processes, cues, lexical entries, articulatory forms, auditory forms, and syntax, one must spend a great deal of time and effort at it.

Many researchers have been studying the effects of motivation and attitude on second language learning because of their drastic effect on the effort spent attending to the learning. If you do not have any interest in learning a new language then you will likely not find yourself able to gain the skills necessary to become a proficient user of that language. Much of this motivation and general influence comes from social pressures and global social norms. If a person is thrust into a multicultural area such as that in many areas of Canada then one will likely learn much of the languages around them. In such situations a person may become bilingual, may find drastic mixing of languages, or may find social bias for using certain forms of language over others.

Even if one commits oneself to the new language, you are still faced with those hurdles. It can be quite a daunting task to put to memory the numerous new lexical entries which the language will surely have (even if there is considerable similarity as with French and Spanish). Cross-linguistically researchers also investigate priming of the two types mentioned earlier in the monolingual studies. In this case we would find phonetic priming of words like ‘flap’ and ‘frappe’ and semantic priming of words like ‘window’ and ‘fenetre’ (if my extremely faded four years of French serves me right!). What is interesting is that once a person has become bilingual (or close) they will never be able to be as fast on some typical lexical identification tasks as a native monolingual. This is evidence of the great effort it takes to master a second language. You are putting your brain through so much more than a person who only must worry about one language that you can not possibly keep up. Though these slow downs are on an incredibly small scale, they are still measurable.

Another interesting consequence of second language learning is the crossing over, or transfer, of syntactic forms. It has been found that some bilinguals will allow syntactic structures from one language guide their utterances in the other. One example would be using the English SVO word order when speaking Italian where it is not important. This leads me to an interesting fact that is beginning to surface. Even an accomplished bilingual who would easily be rated as a native speaker of either language by a native monolingual speaker of that language does not fully learn and use the cue structure of the second language. In other words, a bilingual will still use their native cue structure for their second language even though it may not apply correctly. This is further evidence of just how hard it is to overcome the hurdles presented to second language learning by your native language.

One last interesting fact about bilinguals is that they are found to have varying degrees of separation between the two languages. Some bilinguals are quite ‘balanced’ and keep the two languages quite separate. By this I mean that they do not have as much crossing over between languages. This also usually results in slower processing of language in comparison to less balanced bilinguals and again to monolinguals. This shows an interesting scale of speed on the mixed bilingual side and clean correct speech on the balanced bilingual side. Both have advantages in different situations and both show the difficulty in grafting a second language on top of an existing linguistic framework.

To conclude, language consists of many aspects which include: lexical items, auditory, articulatory, and semantic forms of lexical items, and syntactic structure. We spend a great deal of our lives learning these features, both how to use and how to recognize them. There is a huge number of interesting areas of research contained within each, many of which are just barely beginning to show concrete results. The learning of one language seems to be a daunting task, but yet we handle it with amazing ease. We see, with the learning of a second language, just how difficult it really is. Does this show that there is a critical learning period? Does this show it is innate? All are unanswered questions, and all are flourishing areas for debate. At its root, language is a means for the communication of thoughts. I can only hope that this summary has communicated to you a good overall conception of a portion of the field of linguistics.

References

Bates, E., MacWhinney, B. eds. “The Crosslinguistic Study of Sentence Processing”, Cambridge University Press, 1989

Chomsky, N. “Language and Mind”, Harcourt, Brace, and World, Inc., 1968

de Groot, A. M. B., Barry, C. eds. “The Multilingual Community: Bilingualism”, Lawrence Erlbaum Associates, Publishers, 1992

Nakamura, G. V., Medin, D. L., Taraban, R. eds. “Categorization by Humans and Machines”, Academic Press, Inc., 1993

Osherson, D. N., Lasnik, H. eds. “An Invitation To Cognitive Science - Volume 1 - Language”, The MIT Press, 1990

Reynolds, A. G. ed. “Bilingualism, Multiculturalism, and Second Language Learning”, Lawrence Erlbaum Associates, Publishers, 1991