The clinical utility of linguistics and linguistic innovation is the tools which are provided. The first of these tools are from the history of the two fields and the first personalities involved, the Reverend William Holder, John Thelwall, and Alexander Melville Bell. The current and greater utility is from the theoretical and empirical research pioneered by Noam Chomsky in the 1950s and now developed by a huge worldwide community of scholarship working in almost all fields of linguistics, now dominant in many of these fields, but not in clinical linguistics.

It is important to recognise that Chomsky’s tradition of linguistics started long BEFORE the structuralist tradition represented by William Croft (2002, 2004, 2010), Ben Ambridge (2004, 2010) and others which often seek to represent Chomsky’s tradition as some sort of interloper. Over the past 300 years, the history is the other way round.

Holder would seem to have been the first ever speech and language pathologist. As a founder fellow of the Royal Society, in 1667 he gave what was seemingly an all day seminar to his fellow fellows on his treatement of a deaf boy, Alexander Popham. This was seemingly the first ever therapeutic or educational approach to deafness. The Royal Societyhasd just been founded by Charles ll. It quickly became the world center of advanced science. The fellows were gifted, well-informed intellectuals who met weekly in London to discuss matters of science and mathematics. Charles made sure that the proceedings were promptly reported back to him. He wanted to avoid the fate of his father. The Pophams were a prominent political family. From Holder’s clear description, his treatment was based on a fore-runner version of what is now known as ‘generative phonology’, now a significant field in linguistics. In the 17th century the term, ‘linguistics’ had yet to come into use or existence. It is therefore, also important to notice that it was the clinical impulse which originally inspired the first inklings of generative reasoning. Holder was followed in this way of thinking and clinical practice first by John Thelwall the first half of the 19th century and then by Alexander Bell in the second half. The thinking came before the technical terminology. The useful and positive legacy from these three scholars was lost, deliberately, it would seem. But it can be recovered.

The history

Holder’s originality was to propose that ‘letters’, or speech sounds, or ‘phonemes’ as we would now say, should be viewed, not as such, but as the effect of particular combinations of hierarchically organised ‘matters’, defined by where the tongue is in the mouth, how the airflow is shaped, whether the airflow is allowed to pass through the nose, what is going on in the voice-box or larynx. This happens in what Holder calls the ‘tract of Speech’, what we now call the ‘vocal tract’, which Holder was the first to conceive as an anatomical entity. Holder’s matters are now known as ‘distinctive features‘. Crucially,  for Holder, phonemes were derived from the features, rather than the other way round. So the features were not accidental properties of phonemes, but definitional. Holder is most insistent about this order of definition. He shows this idea by a matrix with rows and columns, and calls this the ‘true Alphabet of Nature… out of which all languages are made’. He clearly conceived of the matrix as applying to all languages. What was language-specific was the pattern of ‘derivations’ or the pathway from the features to the pronunciation. He notes that “the French write some consonants which they do not pronounce to be Indices of the Derivations of their words.” Holder thus brought the mathematical notion of derivation to the study of speech. IThe theory and terminology is still in use, although Holder’s original contribution is largely unknown.

Holder’s hierarchy was a first step towards what in 1985 Nick Clements would call the ‘geometry of phonological features’. This would become the basis for much current thinking about the sound systems of human language, known as phonology, including the notion of ‘non-linearity’. By this idea, features are not in some randomly ordered stack, but they are organised in a hierarchical structure. The vowels are articulated with the tongue in particular configurations, as a function of their more general place in the mouth. Holder detected that these positions in the mouth are organised. This can easily be told by tapping the larynx while forming the mouth into the positons for the vowels in he, hay, har, hoar, who. One of what are now known as the ‘formants’ are organised in this simple linear series. Isaac Newton, another Royal Society fellow, was evidently in Holder’s audience. He later repeated Holder’s point about the organisation of the vowels as a series.

Applying to clinical practice this notion of hierarchically structured speech sounds, Holder would ask Alexander to repeat a series of minimally different forms which did not mean anything, but which were nevertheless possible words. “When you require one vowel of him, he will sometimes stumble on another… And when you have made him perfect at Syllables, then you may reckon that you have taught him all pronunciation of Language, since all words are onely some of these Syllables, or else Syllables compounded of these.”

Holder missed out what is now often referred to as the metricality of ‘word stress’ and the effects of different grammatical roles. These would be for later scholars from John Thelwall to Chomsky and Halle to discover.

Not knowing of Holder’s work, in 1983 I developed the same notion of giving children ‘pretend words’ to say in a highly structured sequence, but now with the natural stress contours of English, as described here under the heading of Possible words.

John Thelwall was the first to distinguish between speech disorders which were, to at least some degree, the effect of a child’s experiences good or bad, and those which were not. In relation to the latter, using the metaphor of musical ‘cadences’, Thelwall developed an early version of metrical theory to describe the characteristic English alternation between stressed and unstressed syllables, applying to all words in the language.

Making Holder’s theory more complete, Alexander Melville Bell showed in a series of books starting in 1849 that English vowels could be defined, not just by the position of the tongue in the mouth, as Holder had proposed, but also by the rounding of the lips.

This lip-rounding relates to the way most English children say the L in words like little and middle, often saying little as LIKU and middle as MIGU. This seems to be conditioned by the difference in stress between the first and second syllables, and to the derivational effect of the fact that the underlying representation of the L sound is with the tongue tip. Next to the tongue tip L the T or D loses its tongue tip articulation, becoming a K and G, by what is known as ‘dissimilation’. (Here I am extending and developing a theory known as ‘Radical Underspecification’ from Diana Archangeli in 1984). Competently spoken  little varies quite widely from one variety of the language to another.

Bell was the first scholar to reflect upon the fact that words like should and have can lose much of their sound structure in sentences like “I should have done that” with the vowels in should and have going unpronounced.

Recent lessons from linguistics

The most profound clinically relevant lesson from linguistics is that there is such a thing as the learnability space. Some aspects of speech and language fall within it, and have to be learnt during every normal childhood. Others fall outside it.

There are two interlinked aspects of what has to be learnt. By one, the grammar has to specify all and only those phenomena that contribute to the faculty of language. The grammar has to do so in such a way that it is finitely learnable, and, by recent formulations, in such a way that it could plausibly have evolved. By this finite learnablity, the learner learns that the order of adjectives like good or bad and what they modify varies acccording to whether it is a pronoun like something, nothing, anything or everything which gets modified or a noun. And this learning is robustly expected. A speaker saying “I see bad nothing” or “I like books good” would be barely understood, if at all. By the second, the grammmar has to be broken down into the smallest possible chunks. Such a grammar is both reductionist and ‘generative’ in the sense that it generates structures freely in two directions, by the speaker creating meaningful forms and by the listener assigning analyses to them. By Chomsky’s seminal 1957 proposal, entitled Syntactic Structures, this breaking down was into two components, a ‘phrase structure’ component and a ‘transformational’ component. Within the latter, the rules manupulated variables commonly shown by numbers. These could be added to, deleted, or reordered, giving what are recognisable as questions, negations, passive structures and so on, with all the elements minutely defined. These elements included the traditional category of tense, as in “She fell over” in contrast to “She falls over”. The novelty here was that all components were defined abstractly. The great empirical advantage of this abstraction is that it makes description more precise. Some like Ambridge sneer at this on the grounds that it is ‘algebraic’,language far more precisely than by the traditional terminologies of classical, renaissance, and early modern scholars. Without this precision, the finite learnability would not be possible.

Phonology

The clearest linguistic contribution to clinical practice is with respect to phonology. Not only is it possible and useful to dig down underneath speech sounds  to the world of features of various sorts, continuance, different articulators, the use of the nasal resonator, and more. It is also possible and useful to take account of the larger function of foot structure or metricality in words like hippopotamus which can cause enduring problems, even for some adults. Issues arise by the complex and little understood interactions between segmental features and metricality. These are partially addressed by the therapy I refer to here as Possible Words. This would not be possible without the singular advances in the study of both segmental and metrical phonology in Noam Chomsky’s and Morris Halle’s monumental and seminal Sound Pattern of English or SPE in 1968.

Syntax

The relations between subjects, verbs, and objects are quite complex, sometimes encoded in metaphor. In some cases the meaning of the verb depends on its relation with subjects and objects, mainly the latter. In “I just killed the wasp” and “I just killed a steak and kidney pie”, the verb kill means two quite different things. In the second it is a metaphor only in the very extended sense that the pie no longer exists as a pie. But in “The steak and kidney pie just killed the patient” the meaning is entirely literal if the patient was almost at the point of death from starvation or if the pie was poisoned or if the patient was critically allergic to kidneys. Most relations between subjects, verbs, and objects are simpler than this. In some languages, the special syntactic structure of the verb is implemented by forms which are more like adjectives, a bit like “The steak and kidney pie was fatal to the patient” without needing a preposition like to. Russian is a well-known case. It uses the adjectival forms only in the past. The way this works, whether in an English-type way or a Russian-past-type way, is something the child learner has to learn. Some languages use only the adjectival forms, and have nothing truly equivalent to verbs, with forms like kill, kills, killed, killing, in an English-type language. This falls within the learnability space.

Much of what falls within the learnability space is mediated in part by universals such as questioning, negation or by the shifting of topic and focus. All of these things are universal. But the way they work is highly language-specific, simple, easy to learn and clear in some languages, less so in others. Some languages mark negation by a form at the beginning of the sentence. English marks negation by the difference between “She’s coming” and “She isn’t coming” with not immediately after the form carrying the tense, present or past, in this case present is. Not simple at all.

Most clinical work on what is often known as ‘expressive’ or ‘productive’ language is focused on single words and the syntax of subjects, verbs and objects, and some simple aspects of phenomena such as negation, as by “The bird is not in the nest” or “The bird is not in the tree” as opposed to “The bird is in the nest” or “The bird is in the tree”. But this extends only to those aspects of grammar normally being learnt between the first and second birthdays. The typically developing two year old is typically very hard to understand other than to those who know him or her well. And the range of what he or she can say and understand is similarly very limited. Somebody with a grammar extending only to this point would be very limited in what he or she could do, not being able to be trusted with anything as dangerous to him or herself or anyone else as a hammer, a saw or a sharp knife. It would be impossible to explain the potential dangers with sufficient clarity.

English mostly marks shifts of focus or topic by the device known as ‘passivisation’ as by “An asteroid ended the age of the dinosaurs” passivised and questioned in “The age of the dinosaurs was ended? I don’t think so. Modern birds are just highly evolved miniature dinosaurs with forelimbs which are now used as wings or flippers and feathers to keep them warm.” The way that English does passivisation is highly complex, involving the order of the words, the structure of the verb itself with either -en or -ed on the right edge, a special use of the be verb, and the original subject turned into a by phrase. Unsurprisingly this takes a long time to learn, especially when this interacts with questioning and negation as in “Wasn’t the age of the dinosaurs ended by an asteroid?”

Much simpler forms are common in school. But children with language issues, with delayed or disordered language development, often have great difficulties with very common passive forms. In “You were given a book. Where is it?” you were given a book is what is often called a ‘short passive’ – with no by phrase. But if the passive structure is not correctly analysed and understood, the follow up question is incomprehensible.

Consider the sentence, “Give me the toy you had yesterday.” The toy may be just out of the reach of the adult saying it, but within the reach of the child to whom it is addressed. Said the day after a birthday, the reference might plausibly be to a much wanted present the child was given the day before. The child may respond appropriately. But this does not mean that the child has properly understood the sentence. In “Give me the toy you had yesterday” you had yesterday is a what is known as a ‘relative clause’. It restricts the reference down from all the toys in the world or all the toys to which the child has access to one particular toy. So relative clauses are very useful conversational devices. English happens to use them very widely, to both subjects, objects, what are known as ‘indirect objects’ as in “The child you gave a toy to yesterday is coming back to get it”, and it allows the form which introduces the form which introduces relativisation, which or who or where or that to go unpronounced. Some languages allow only the subjects of sentences to be relativised. In such languages “The toy you had yesterday is upstairs” is grammatical”, but “Give me the toy you had yesterday” may be uninterpretable – either because there is no ‘relativiser’ or because the toy is a grammatical object here, or both. Until Syntactic Structures, English relatives were a very complex aspect of grammar. But by Chomsky’s reductionism, it could be seen that the relative clause just ‘recycles’ the structure of the main clause. This is easiest to see in a form less likely to be addressed to a child because it is potentially confusing, “I use the recipe you use” or “I use the recipe which you use”. By some current models of syntax, which is copied from a position as the object of use to a position where it introduces and defines the relative clause as such, and is then optionally left unpronounced. The basic structure of the two clauses is the same. The difference is in the two subjects, I and you. Not simple, but simpler than non-generative alternatives. The principle here is known as ‘cyclicity’. It  minimises repetition.

The clinical utility here is that the reductionism of generative grammar opens the door to a much wider sort of intervention than possible hitherto. Since 1957 this reductionist impulse has been taken forward in a number of ways, by the notion of Merge, reducing  all syntactic operations to the simplest logically possible formation of a binary set of items, from Chomsky (1995), and the theory of Phase, applying syntactic operations in a series of phases, each doing essentially just one thing, from Chomsky (2000). Both of these innovations have been hugely develped in subsequent work. The net effect is to further widen the scope of clinical interventions.

Semantics

By framework here, children do not need to learn that people and things have names. There is such a thing as reference. others have minds of their own. To a much lesser degree than humans, various non-humans sometimes display a similar understanding. What is commonly called the ‘theory of mind’ or TOM is a human endowment. Michael Tomasello, whose rejection of generative linguistics is rejected here, is a leading TOM researcher and scholar. Similarly there a robust, neurophysiologically well-defined cognitive functionality allowing humans at least and probably some non-humans to navigate familiar spaces by touch if the light fails completely, and to navigate less familiar spaces by memory in unfamiliar ways if one path is blocked. Small children often love games of hide and seek in which this functionality is exploited. It can be damaged by neurological accidents. It may be that it is related to the core notion of surface on which locative relations and prepositions seem to be defined. The objective world is partly defined by the relations that hold within it. Houses have roofs, pointed if there is rain, chimneys if they need to be heated, windows in the middle of walls, and doors the height of an adult human going down to ground or floor level. Separately there is also sequence and causality. Eggs have to be broken to fry them or to make an omelette. The world of propositions is partly defined by truths and untruths. Accidentally or otherwise, not everything that is said is true. Similarly, conversation works partly o n the basis that not everything is known or understood. Sometimes questions need to be asked. Or commands need to be given, sometimes to preserve human life. And roles in conversation change from one item of dialogue to the next. In “I didn’t say that” and “You most certainly did”, I and you may refer to the same individual. None of these things vary from language to language. They do not or should not need to be taught. If any of these things do need to be taught, this is a different sort of therapeutic or educational enterprise from work on phonology or syntax.