1

2

3

Frisch, Karl von. 1967. The Dance Language and Orientation of Bees. Cambridge, Mass.: The Belknap Press of Harvard University Press.

Traffic lights / clothes

Picture of gesture in ASL for “I Love You”

4

Definition rests on spoken or written but what about signed?

Definition as rests on identifying the medium (written or spoken) of phenomenon notthe phenomenon itself. It is like saying “music is what gets played on the radio”.

Channel v. Medium: medium = e.g. 'phonic', channel = e.g. 'vocal auditory'. Possiblefor the former to be transmitted through various channels, e.g. sound waves,phonetic script, vibrations etc.

5

6

“Except for a small number of clichéd formulas, just about any sentence that youproduce or understand is a brand new combination produced for the first timeperhaps in your life, perhaps even in the history of the species.”

Steven Pinker (2012): Linguistics as a Window to Understanding the Mind http://www.floatinguniversity.com/pinker-transcript

7

Apparently, the longest sentence in English Literature is by William Faulkener in thenovel Absalom, Absalom! According to the Guinness Book of Records, 1,287 wordslong. James Joyce, in Ulysses claimed to have written one 4,391 word long.

8

A second insight is that languages have a syntax which can’t be identified with theirmeaning. Now, the only quotation that I know of, of a linguist that has actually madeit into Bartlett’s Familiar Quotations, is the following sentence from Chomsky, from1956, “Colorless, green ideas sleep furiously.” Well, what’s the point of thatsentence? The point is that it is very close to meaningless. On the other hand, anyEnglish speaker can instantly recognize that it conforms to the patterns of Englishsyntax. Compare, for example, “furiously sleep ideas dream colorless,” which is alsomeaningless, but we perceive as a word salad.” Pinker (2012).

”Buffalo buffalo Buffalo buffalo buffalo buffalo Buffalo buffalo” - "Bison from Buffalo,that bison from Buffalo bully, themselves bully bison from Buffalo.“ (Borgmann 1967).This an interesting example of how the rules of syntax can allow one to construct(and interpret) a string made of the same word repeated seven [sic] times.

9

Linguistic structures are not organised not a simple linear fashion but as hierarchicalstructures and dependencies between items operate within the structure of the treenot between words in the order in which they occur in the so-called surfacestructure: “Either ….. if……”; “If …. then……”, “Neither ….. nor.”

Also it is much easier to explain the diffeence between two sentences like: "I sawJohn" and "Who did you see?"

in terms of transformations of the same (kernal) structure rather that two seperatelygenerated sentences. Transformations work most logically with tree-like "deep"structures than they do with strings.

10

The concepts of deep and surface structure are no longer widely used by Chomskyansbut they are still useful in understanding how the sentences that are actuallyproduced are merely the end-products of processes occurring out of sight in the mindof the speaker, like roots hidden by the earth underneath a tree.

In example, a variety of nested dependencies.

11

This is because Chomsky is ultimately interested in how the mind works, not howhumans actually communicate with each other.

12

Chomsky concentrates on language as an abstract system operating within thehuman mind and plays no attention to its social function or to particular contexts ofits use.

See Halliday, Michael A. K. (1978) Language as social semiotic: The social interpretation of language and meaning. Maryland: University Park Press (p. 2).

13

According to Halliday, there are three functional components of meaning as conveyedby grammar: the ideational (the transitivity structure of clauses, which expresses therepresentational meaning); the interpersonal (the mood structure which expressesinteractional meaning); and the textual (the theme structure that expresses how themessage is organised, analogous to the information structure of functional linguists).Halliday notes that, at the ideational level, the clause, “as message” as opposed to“as an exchange” or “as an organised unit”, is a representation of a process, in verygeneral terms, some action, event, happening or situation. This process, for him, is initself the most basic kind of “something” to be represented.

The propositional content of a clause is only relevant at the interpersonal level,where the clause is an exchange either of information (“The present King of France isbald”) or of goods and services, where the commodity exchanged is non-verbal (e.g.“Kiss me!”; “Pass the salt”; - Halliday 1985: 68).

14

Representation: the ideational elements are the participants in the process. The verbbeing the central element. There are different kinds of processes, each with its ownset of participants.

Exchange: the interpersonal elements are structured around the subject and modal, ifpresent, (the mood) and the residue (basically the predicate). it is the truth of themood that gives value to the exchange (e.g. “I love you” / “I don’t love you”.)

Message: the textual elements relate to the information structure of the clause.Which bits of information are presented (first) and function as the starting point forthe message and which are given last (the rheme). In this example “I” is giveninformation and thus low in “communicative dynamism” while “love you” representsthe new, “communicatively dynamic”, information.

Halliday is quite literal in his interpretation of first and last, sticking rigidly to wordorder. Other linguists, e.g. Firbas (1992), argue that theme constitutes least“communicatively dynamic” information and does not always have to come first inthe sentence but, wherever it comes, constitutes the psychological starting point forthe message.

15

PROCESSES and PARTICIPANT ROLES

Relational intensive - identifying “Tom (Identified) is the leader

(Identifier)” - attributive “Sarah (Carrier) is wise (Attribute)”

circumstantial – identifying “Tomorrow (Identified) is the 10th

(Identifier)”- attributive “The fair (Carrier) is on Tuesday

(Attribute)”possessive - identifying “Peter (Possessor) has a piano

(Possessed)”- attributive “The piano (Carrier) is Peter’s

(Attribute)”Verbal “John (Sayer) said ‘I’m hungry’ (Quoted / Reported)”Mental “Children (Senser) fear ghosts (Phenomenon)”Behavioural “The mock turtle (Behaver) signed”Material “The lion (Actor) caught the tourist (Goal / Patient)”Existential “There was a storm (Existent)”

15

16

Different languages are structured in different ways and the degree of correspondence between anytwo may be variable. For example, Italian is a largely inflectional language and consequently it allowspronoun dropping (the ellipsis of subject pronouns) and has a fairly flexible word order. English islargely isolating (there are few inflections and the function of the word is shown by its position in thestructure). Consequently pronouns cannot be dropped and word order is more rigid.

The so-called mould theory of language as opposed to the cloak theory. The former is associated withthe Whorf hypothesis - In brief, according to this theory ones L1 determines the way one things. Thelatter is associated with the more classical idea that language is merely the dress for thoughts, i.e.what one thinks determines how one uses language.

Human functions, such as thinking, speaking and understanding, were the major points thatpsychologists, sociologists or anthropologists were focused. Thus, within linguistic theory, to extremepositions has been developed in order to determine the relationship between language and thought:"mould theories" and "cloak theories". Mould theory indicates that language is the result of thoughtand cloaks theory reveals that language is the cloak that adapting to the usual categories of thespeaker's thoughts.

The Sapir - Whorf hypothesis, named after the Edward Sapir and Benjamin Lee Whorf,is a mould theory. Sapir, who was a linguist and anthropologist and Whorf, who waslinguist, were interested in comparing languages, which are one of the majorcharacteristics of a culture. Individuals that are born in a particular community andhave a particular language, it is obvious that the language will be the greatestinfluence for them. That is because it is used by the most people, so becomes ‘public’and not private as are the individual’s 'thoughts'.

Sapir and Whorf argue that vocabulary determines the categories that are used toperceive and understand the world. For example, Hopi Indians have one word for"insect", "aeroplane", and "pilot". Also Zuni Indians do not distinguish, verbally,between yellow and orange. According to Sapir and Whorf, grammar influences theway that we think and perceive the world, as well. For example, in the Hopi languagethey are no distinctions between the past, present and future.

According to Sapir, we see and experience very largely because of our language habitsand this underlines our access to the real worlds meaning that people who speakdifferent languages perceive and think about the world quite differently (the linguisticrelatively hypothesis). His student, Whorf, extended this position, arguing thatthought is determined by language (linguistic determinism), according to which,people who speak different languages have also a different view of the world, adifferent way of perceiving things and generally thinking. The world is different

17

according to what language we speak or to what language we 'think'.

17

Language of Thought (LOT) would constitute a symbolic system of atomic particles(basic ideas) that have semantic and syntactic properties that combine in acomputational system to form a potentially infinite variety of structures e.g. thethought Mary loves John would be realized in Mentalese by separate elements ortokens (ultimately as some kind of physical form e.g. as neutrophysiologicalproperties in the brain or central nervous system), corresponding, one wouldpresume, more or less to the linguistic elements Mary, Love and John that wouldcombine together or with other items in a variety of ways to express other thoughts:John loves Mary, Mary loves Fred, Fred hates Mary etc. That is, for at least"propositional attitudes" ("Propositional attitudes are thoughts described bysuch ..forms as "S believes that P", 'S hopes that P" etc" - The Stanford Encyclopediaof Philosophy - online edition). LOT and the accompanying Representational Theoryof Mind is considered a working hypothesis in cognitive science (see for exampleField 1978, Fodor 1987, Fodor and Pylyshyn 1988). The online Stanford Encyclopediaof Philosophy .article on the Language of Thought, succinctly states that"understanding a sentence is to entertain the thought proposition it expresses."

However, the existence of Mentalese is only a hypothesis, even if a strong one.Behaviourists have always argued that language is thought and more recently the socalled connectivists have argued for a kind of associationalism in its place (wherethoughts are determined by behaviourist factors and can be seen as responses tospecific stimuli (at least originally external) and not part of a system which could, intheory, be entirely internal - see Smolensky 1990, Chalmers 1993. Even if it does exist,it is not certain that in the near future, or even longer, humankind will be able toaccess it directly through artificial technological means (e.g. some form of brainscanning). Furthermore even if such a thing were possible, there is no guarantee thatthe information thus collected would help much - as the saying goes: "If our brains

18

were simple enough to be understood, we'd be so simple that we couldn't" (thisversion of this oft-cited quote is from Wikiquote which lists it as in need of a source).

Supporters of the LOT hypothesis, i.e. cognitive scientists, would argue that basicthought patterns and the way they were represented in Mentalese would be innatethat is an inherent feature of the species. This is a relatively modern view and verymuch in line with the view of linguists such as Chomsky who have argued that therules necessary for linguistic competence are innate.. Previously, it was assumed thatbabies were born with a mind that was like a blank slate. language was somehowleant by imitation.

18

19

Language probably evolved somewhere in the region of 50-150,000 thousand yearsago (based on so-called to behavioural modernity: artwork, tool use, ritual) right backto humans’ earliest homo ancestors and cousins, the so-called hominins (2.3-2.4million years ago) even perhaps all the way back to the common ancestor forhominins and its sister pan genus (i.e. chimpanzees and bonobos), 5-6 million yearsago. As can be seen, there is no agreement on precise time or on the sequence ofevents.

It is difficult to tell as fossils don’t leave traces of the brain and it is largely guessworkwhether a particular artefact was made by someone capable of speech or not.However, the advent of highly organised (hunter-gatherer and then agricultural)societies relying on cooperation, not to mention the beginnings of artwork, as well ascertain physical features of humans (such as the dropping of the larynx) point to thefact that language happened earlier rather than later in human history.

Certainly, the development of writing systems happened much, much later. The firstwriting systems and alphabets were only introduced 4,000 years ago and, indeed,even today most languages don’t have their own alphabets (Old English was notwritten down extensively until the Roman alphabet was adopted), so writing is clearlynot an intrinsic part of language.

20

Seeing that all human societies ever studied have and use language, and seeing that as an facility it most likely evolved once and in one place, language must date back to the first groups of humans living in the same geographical area (probably East Africa near the Serengeti plain), some time before the great migrations (circa 100,000 years ago).

21

Max Müller (1861) famously reduced the various then existing theories to fourapproaches with distinctive names: the bow-wow or cuckoo theory (attributed toJohann Gottfried Herder, who Darwin seemed to agree with) that sees language asderived from sounds made in imitation of noises produced by animals, especiallypredators; the pooh-pooh theory where words derive from instinctive exclamationsand other unconscious affective displays (e.g. exclamations, laughter); the ding-dongtheory claiming that objects have a vibrating natural resonance which wasrepresented somehow in the first words; and finally the yo-he-ho theory, wherebylanguage derived from rhythmic sounds used to coordinate collective physical effort.Such theories remain sketchy and largely conjectural in nature. According to the ta-tatheory, which was not listed by Müller's, as it was put forward by Paget (1930),humans starting using the tongue to mimic manual gestures, rendering them audible.

22

The mother tongue hypothesis rests on the fact that language requires trust andshared interests as it can easily be used to deceive, so to evolve it would have to bebeneficial to all using it. Fitch (2004) argued that language developed betweenpeople with shared genetic interests (i.e. kin, principally mothers and their off-spring), who could trust each other and use language to ensure each other’s survival.

Again to deal with the problem of trust, Ulbæk (1998) puts forward the idea ofobligatory reciprocal altruism: the Darwinian principle of reciprocal altruism thatexists between individuals that habitually co-operate. For language to have evolved itwould have required strict adherence to this principle within the whole social group,implying some kind of regulation and enforcement.

On a different track, Dunbar (1996) argues that language, in particular gossip, evolvedto perform the function of grooming (and generally displaying commitment) whichbecame necessary as human groups got larger and larger, making it harder todedicate the necessary time to physically groom each other. Language appeared firstas vocal grooming and then vocal language (i.e. gossip).

Ritual/speech co-evolution: among others, Rappaport (1999) argues that languagedid not evolve separately but as part and parcel of human symbolicculture. Language could not work outside a specific context of social mechanismsand institutions, like credit cards without a banking system.

23

Tower of babel hypothesis (Pagel 2012): language evolved develop to excludeoutsiders and form break away groups.

Gestural theory: (see Jastrow 1886, Pinker and Bloom 1990) argues that languagedeveloped out, in an example of an adaptation, of gestures, and many apesinstinctively use gestures similar to these of humans. Cognitive scientists have foundthat the gestures and language use similar neural systems. Humans may have usedvocal calls as substitutes for gestures as they hands became increasingly occupiedwith other tasks (e.g. tool use).

Falk (2004), putting the baby down theory, as humans lost body hair, children had noway to cling on to mothers, and if mother’s carried them, they were unable to usetheir hands. Language developed as “motherese” a means to communicate withbabies who were left on the ground while the mother worked and in this way, bothreassure them, and be able to check upon and control them.

Of course, language may have evolved for a variety of reasons. It could have emergedin one context and then, by chance, proved useful in others and thus evolved in anew direction in the same way that a piece of technology for one use may be“hacked” and serve other purposes (e.g. a plastic bottle that is used as a phoneholder).

23

Contemporary anthropologists, evolutionary biologists, and cognitive scientists haveput forward more detailed theories regarding how, when and why language cameabout – many of these highly technical regarding mind design and cognition or withinthe field of artificial intelligence, the use of computer simulations – and it hasemerged that question is in all probability too complex to have a single answer andindeed it should be borne in mind that language, by its nature, a multifacetedphenomenon must have come about for a variety of reasons and by a variety ofdifferent avenues.

One of the major concerns which such theories have to deal with is whetherlanguage can be seen as something quiet distinct and unique to humans (an ideashared by both Max Müller and Chomsky) or whether it can be seen as a albeit highlysophisticated continuation of forms of animal communication (Pinker). The so-calledcontinuity theory would have it that, as a result of what biologists call punctuatedevolution, there was some dramatic break-off point where some change, perhaps initself quiet minor but with disproportional effects much like the so-called butterflyeffect in chaos theory (see Lorenz 1963), rapidly brought about a language facultywhich immediately singled out humans or other so-called hominins (that is Homosapiens and other, now all extinct, members of the genus Homo: Homo habilis, Homoerectus, or Homo neanderthalensis etc.) as something special in respect to their apecousins (i.e. other members of the Hominoidea superfamily); by contrast thecontinuity theory requires a gradual set of largely uniform changes, largely as

24

adaptions of existing facilities, (Pinker and Bloom 1990) with no specific dawn of thelanguage faculty.

See for example: Functional Magnetic Imaging Resonance (fMRI) or Neuroimagingstudies highlighting activity in different areas of the brain while using language andcomparing it to studies of other primates and examining, for instance, how thelanguage faculty may have evolved out of so-called mirror neurons (see Skoyles 2000), the ability to interpret, understand, and to mimic gestures (which some primatesinstinctively do), for example, or out of the faculty to see things, develop hand-eyecoordination (vital for primate’s swinging at speed through the trees) and visualisethings in “the mind’s eye” and the development in primates of “two and halfdimension” vision (see, for example, Pinker 1997: 211-298).

See, for example, the work of Steel (2011) who within the field of Fluid ConstructionGrammar, uses robotic agents to ascertain from interaction between robotic agentswhat precise cognitive abilities are necessary for the evolution of specific aspects oflanguage.

Two palaeontologists, Eldredge and Gould (1972), attack the perceived gradualevolution or phyletic gradualism that they identify with Darwin, claiming that thefossil records support instead a punctuated equilibrium. This latter idea, has also beentaken up by some exponents of so-called intelligent design, who argue thatintervention of some divine force can be seen in evolution, and thus that creationismand evolution can be reconciled. Such views are widely rejected by evolutionarybiologists even though punctuated equilibrium continues to gain traction and somelike Dawkins (1996), who points out that Darwin himself never said that all evolutionhad to be uniform and gradual. There are of course several other theories ofevolution including punctuated gradualism (Malmgren et al. 1983) where gradualchange is interspersed with periods of sudden change, and multi-tempoed theory ofevolutionary change (Simpson: 1944) where different taxonomic groups or species(e.g. Homo sapiens, Homo habilis, or Homo erectus) may evolve at radically differenttempos with so-called quantum evolution constituting a major shift, especially at thelevel of higher taxonomic categories such as genus (e.g. Homo, which includes homosapiens and its now fellow so-called hominins such as Homo habilis, Homo erectus, orHomo neanderthalensis).

24

25

Codes rely on their being interpretants. According to C.S. Pierce (1931) a sign has three parts:

the Representamen: the form which the sign takes; an Interpretant: not aninterpreter but rather the sense made of the sign; an Object: to which the sign refers.Semiotics however can define sign too widely and fail to distinguish between signsand non-signs. In essence a sign is anything that can be interpreted as a sign (but seeSperber and Wilson’s concept of ostension below).

Shannon and Weaver’s so-called mathematical model of communication assumesthat a message can be passed along as an immutable unit. It doesn’t take intoaccount incomplete or divergent interpretations.

26

Traditionally, language is seen as resting exclusively on coding – e.g. Saussure (1916), Shannon andWeaver (1949). However, many now think that inference – deduction -- is more fundamental. Indeedthey argue that coding itself rests on inference: for code to be effective both people have to recogniseeach other’s intention to communicate. As a prerequisite for coding, this can’t actually be encoded buthas to be deduced from the context, from the other person’s behaviour.

Example 1 is an example of a simple code. The idea that Socrates is mortal is expressed explicitly bythe sentence itself.

Example 2 is at one level a code at another inference because the information that “Socrates ismortal” is explicit: it is not actually encoded in a single sentence as such, but deducible from puttingthe two sentences together.

Example 3 (from Brown and Yule 1983) is pure inference. The messages “Open the door” – “No, Ican’t” are not encoded anywhere in the message. Searle (1969/75) argued that this was a so-called“indirect speech act”, a sophisticated type of code. Grice and Sperber and Wilson (see next slide)provided a simpler, much more robust explanation using the concept of inference.

27

Grice 1975 – Conversational Maxims (see also the so-called Co-operative principle)

For Sperber and Wilson reduce Grice’s principles to one: “Be relevant”. For them,communication is not a “unitary phenomenon to be described by a single model”(1986: 24); it relies on a combination of both coding and inference. They say (1986:9), “It is true that a language is a code which pairs phonetic and semanticrepresentations of sentences. However, there is a gap between the semanticrepresentations of sentences and the thoughts actually communicated by utterances.This gap is filled not by more coding, but by inference.” Indeed, for Sperber andWilson, inference is more than just a gap-filler; it is the most basic kind ofcommunication , and thus, instead of being a supplement to coding, it is coding, infact, which supplements inference; “Verbal communication is a complex form ofcommunication. Linguistic coding and decoding is involved, but the linguistic meaningof an uttered sentence falls short of encoding what the speaker means: it merelyhelps the audience infer what she means. The output of decoding is correctly treatedby the audience as a piece of evidence about the communicator’s intentions. In otherwords, a coding-decoding process is subservient to a Gricean inferential process.”(1986: 27).

28

It is well known that animals can achieve highly complex things like web or nestbuilding by instinct, but traditionally may scholars have been very wary aboutattributing language to the same kind of instinct. Perhaps this is because, until, and ora long time after, Darwin, people were uncomfortable about comparing humans toanimals (and all the moral and ethical issues that this raises), and language wouldseem to be the intrinsic human activity par excellence. The one thing that clearlydistinguishes humans from animals (see Palmer 1984, among others, who arguedthat Homo loquens would have been a better name for the species Homo sapiens).

See Edward Sapir (1921: 3-4) “The process of acquiring speech is, in sober fact, anutterly different sort of thing from the process of learning to walk. In the case of thelatter function, culture, in other words, the traditional body of social usage, is notseriously brought into play. The child is individually equipped, by the complex set offactors that we term biological heredity, to make all the needed muscular andnervous adjustments that result in walking. Indeed, the very conformation of thesemuscles and of the appropriate parts of the nervous system may be said to beprimarily adapted to the movements made in walking and in similar activities. In avery real sense the normal human being is predestined to walk, not because hiselders will assist him to learn the art, but because his organism is prepared frombirth, or even from the moment of conception, to take on all those expenditures ofnervous energy and all those muscular adaptations that result in walking. To put itconcisely, walking is an inherent, biological function of man.

29

Not so language. It is of course true that in a certain sense the individual ispredestined to talk, but that is due entirely to the circumstance that he is born notmerely in nature, but in the lap of a society that is certain, reasonably certain, to leadhim to its traditions. Eliminate society and there is every reason to believe that he willlearn to walk, if, indeed, he survives at all. But it is just as certain that he will neverlearn to talk, that is, to communicate ideas according to the traditional system of aparticular society.”

29

It is not the language (e.g. English, Chinese, Arabic, Italian) which is instinctive but the ability to acquire and use language.

Languages do function in similar ways (which are not always superficially obvious) however because they share the same structural principles. It is indeed these principle which constitute what Chomsky has called UG (Universal Grammar).

30

31

In theory if languages were separately evolved or learnt then they could bestructured according to numerous different rationales (e.g soft things versus hardthings).

Quine (1960, 1969), discussing what he termed ontological relativity, argued that, intheory at least, reference is indeterminate, given that one can never be actually surewhat a word refers to. To use his much quoted example, if an anthropologist studyingan unknown language hears a native speaker use the word ‘gavagai’ in the presenceof a rabbit, it would be wrong to deduce from this that it corresponds to the Englishword ‘rabbit’. Among countless others, it could mean “a rabbitting”, “an instance ofrabbithood”, “a rabbit here and now”, “a rabbit in motion” or “separate bits of arabbit joined together”. Many of these would stem from the speakers of the languagein question having a radically different means from a English-speaker’s of perceivingthings in the world and of classifying them.

In Women, Fire and Dangerous Things, Lakoff (1990) reviews a wide range of studiesin "cognitive semantics," a new field that attempts to understand mind throughempirical studies of the way people categorize: subtitle of work: “What CategoriesReveal about the Mind”. He provides several detailed conceptual "case studies,"which aptly bring out the richness of the English language, and Whorfian-typeexaminations of the way different cultures view the world as exemplified in theirlanguages (the book's title derives from a classification in Dyirbal, an Aborigine

32

language of Australia).

32

Leaving aside highly specialised and technical contexts, when it comes to the basicconcepts such as entity and process, animate and inanimate entity, parts and wholesetc, anthropologists and cognitive scientists have argued that, in part, such systems ofclassification are instinctive. One reason that they may be is practical. Pinker (1994)argues that if categorisation were ad hoc, so to speak, the task of learning newvocabulary (something at which children acquiring their first language are remarkablyquick and proficient) would be considerably more difficult. If a child had to considerevery one of the numerous possibilities of meaning identified by Quine, it would behighly unlikely that he or she could learn to speak in only a few years, if ever. The ideathat certain types of classification are instinctive also refutes the Whorfian hypothesis(Whorf 1940) that thought is conditioned by language

Anthropologists such as Sperber (1982) and Brown (1991) have found that peoples inwidely different areas, with widely different degrees of technological and scientificknowledge, tend to share aspects of a metaculture and some apparently universalconcepts. This is shown by the fact that they tend to classify things such as plants,animals, and sundry everyday items in similar ways. For further discussion, see Pinker1994: 419-427.

Pinker (1994: 424-425) reports on some experiments by Keil (1989) into inherentessentialism. Children had to say whether, after various hypothetical transformations,artefacts (man-made objects) and natural kinds (animals and plants) remained the

33

same. They accepted that artefacts could be transformed into things of a differenttype; for example, a coffee pot could be changed into a birdfeeder. However, theyassumed natural kinds to be immutable and to contain some unalterable essence.Thus, even if a racoon was shaved and dyed black and had a white stripe painteddown its back, it was still a racoon, and not a skunk.

33

The distinguishing feature of an ergative language is that it maintains an equivalencebetween the object of a transitive verb and the subject of an intransitive verb, whiletreating the agent of a transitive verb differently. This contrasts with nominative-accusative languages (such as English), where the subject of transitive and intransitiveverbs are treated like each other but distinctly from the object of a transitive verb.

Extension = Transitivity

Actor involved in process. Does that process extend beyond actor or not? "The lion chased the tourist" relates to the lion ‘ran’

Causation = Ergative

A participant is engaged in a process, was process brought about by that participant, or by some other entity?

In this perspective "The lion chased the tourist" relates more to "the Tourist ran" than"the lion ran". This pattern in English become more predominant with verbs like "thetourist woke" and "the lion woke the tourist" being more numerous than the trulytransitive / intransitive "the man hunted" / "the man hunted the lion"

34

Even if basic word order were not a universal (assuming that languages could onlyever be structured along the lines of subjects, verbs and objects – and the existenceof ergative languages that this so, in any case) there are still unexplained patterns inthe word order.

There are six possible orders and if languages did not share common principles thenstatistically one would expect a random, and equal, occurrence of the variouspossible orders (like throwing a dice, one hundred times). In fact, as the graph on theright shows, two of the possible orders: SOV (like Pashto, Latin, Japanese, Afrikaans);and SVO (e.g. English, Hausa, Mandarin, Russian) together account for 87% oflanguages surveyed. This distribution is highly unlikely to be the product of chanceand points to some profound link between languages.

35

Universalism argues that any meaning in one language can be translated into others,and as a result people from different cultures do not think of the world differently. Incontrast to Sapir and Whorf, Vygotsky believes that language and thought are relatedbut that each one operates separately. This is shown by some studies in perception ofcolour. Berlin and Kay (1969) "Some apparent uniformities between languages incolour-naming“: Heider (1972) and Roch (1973) observing the Dani of Indonesia NewGuinea show that, despite important linguistic differences in colour terms betweenDani and English speakers, the Dani could learn arbitrary names for the missing focalcolours (named in English), so that both groups showed better recognition names forfocal colours.

Nowadays it is widely accepted that language can influence thought but it is acceptedthat it does not actually determine it (if language really did precede thought, asPinker 1994 has pointed out, one could not account for language acquisition in thefirst place). Scholars do disagree on the degree of influence of language on thoughtand vice versa. Some would argue that language's influence is strong. On translation,Steiner (1995) takes the stance that the actual verbalization of the message (the lexisused, the type of syntactic structure) affects its meaning and thus that any rewordingof a message involves hermeneutic processes that change its message content. Those,like cognitive scientists, who see the thought process as paramount, argue that thesame thought can be expressed in different ways and thus that rewording whileretaining meaning is possible. This would be due largely to the imperfection and

36

redundancy inherent in natural languages.

36

There may be languages that are exceptional in some way (e.g. English has no 2nd

person pronoun) but such cases are explicable by historical developments (e.g. Middle English did have a 2nd person singular pronoun “thou”).

37

Prior to Chomsky the prevailing view of language acquisition was that children learnttheir language either by mimicry or through instruction.

This view is empiricist (e.g. Locke) in nature and is based on the idea that allknowledge comes from sensory experience. It is associated with behaviourism inpsychology.

Chomsky (1966) adopts a rationalist, Cartesian, “nativist” stance. language is innate.to prove this he shows that language is just too complicated to be learnt by younginfants through instruction or by imitation.

38

1) Children don’t get enough, or the right kind of input to be able to reconstruct andmimic a language from scratch.

Parents often use Motherese and, in any case, there just isn’t the opportunity to hearand learn by heart all the possible structure of a language. It would take a lifetime , ifthat, rather than the typical 5-7 years that it takes a child to become fluent,

Even babies growing up in very deprived, isolated, circumstances show nonethelessan ability to acquire and use language. Abandoned or abused children brought up inisolation have been shown, if rescued young enough, to acquire language, onceexposed to other speakers, at a rapid rate. By contrast efforts to teach chimpanzeesand gorillas to use language have been long and laborious and the results highlyquestionable. There have even been cases of children abandoned together who havedeveloped their own language, e.g. Poto and Cabengo, American identical twins who,having little exposure to spoken language, used an invented language until the age ofabout eight. (see the documentary Poto and Cabengo by Jean-Pierre Gorin (1979)).

Or cases of idioglossia, play or private languages, used typically between two people,often children (“Polari” or “Parlare” is an example of a obscure private language usedby the wider gay community in 1960s London which also made its way,surreptitiously, onto the BBC, but which may have originated much earlier in London’ssubcultures of prostitutes, fish market workers). There is also the documented tragic

39

case of June and Jennifer Gibbons in Wales: two identical twin sisters originally fromBarbados (the “silent twins”) in the 1970s who, as the only black children in theirschool, suffered bullying from the other children and eventually withdrew into theirown world developing a language only comprehensible to them and refused toconverse with anyone else.

39

2) Children might be able to work out simple questions from pairs of sentences like“Is the man tall?” – “Yes the man is tall.” but it is hard to see how they could work outcomplicated sentences like “Is the man who is tall in the room?” rather than *“is theman who tall is in the room” which might seem possible if sentences were structuredas word-strings not trees:

“Well,” Chomsky argues, “if you were actually to look at the kind of language that allof us hear, it’s actually quite rare to hear a sentence like, “Is the man who is tall in theroom? The kind of input that would logically inform you that the word-by-word ruleis wrong and the structure dependent rule is right. Nonetheless, we all grow up intoadults who unconsciously use the structure dependent rule rather than the word-by-word rule. Moreover, children don’t make errors like, “is the man who tall is in theroom,” as soon as they begin to form complex questions, they use the structuredependent rule. And that,” Chomsky argues, “is evidence that structure dependentrules are part of the definition of universal grammar that children are born with.”Steven Pinker (2012).

40

3) Children can be creative with language and create new language which followcomplex grammar rules – which is unlikely they would have been exposed to in orderto mimic or could have been taught. If language was based on nurture and mimicrythen ungrammatical structures should only occur sporadically, instead children usethem consistently.

For example, children go through a long stage in which they make errors like, “Weholded the baby rabbits” or “He teared the paper and then he sticked it.” (examplefrom Pinker 2012) -- cases in which they over generalize the regular rule of formingthe past tense, add ‘ed’ to irregular verbs like “hold,” “stick” or “tear.” And it’s easy toshow… it’s easy to get children to flaunt this ability to apply rules productively in alaboratory demonstration called the “Wug Test”. A child is shown a picture of a littlebird and is told “This is a wug.” The child is shown another picture and told, “Well,now there are two of them.” Children will fill in the gap by saying “wugs.” Again, aform they could not have memorize because it’s invented for the experiment, but itshows that they have productive skills.

Finally the “mice-eater experiment” (Kiparsky 1982, Gordon 1986). Children wereshown a puppet and told "Here is a monster who likes to eat mud. What do you callhim?“ and were given the answer, a mud-eater. Children called a "monster who likesto eat mice," a mice-eater. But a "monster who likes to eat rats" was called only a rat-eater (never a rats-eater) -- Even the children who made the error mouses, never

41

called the puppet a mouses-eater. "The children, in other words, respected the subtlerestrictions on combining plurals and compounds inherent in the word structurerules. This suggests that the rules take the same form in the unconscious mind of thechild as they do in the unconscious mind of the adult.” (Pinker 1994:147)

41

A dropped Larynx has increased the risk of choking. This mutation helps speech andbecome the norm, even if it is potentially dangerous, showing that speech wasimportant for our ancestor’s survival.

The human brain only gets fully-wired about the age of 12 months. If babies wereborn with fully functioning brains, not only would pregnancy last longer but also theirheads would be bigger making childbirth more dangerous for both mother and child.

Children who immigrate and acquire new languages can do so as well as a nativespeaker provided they do so before puberty. After that, they rarely reach native-speaker competence. Pinker (1994: 291) gives the example of Henry Kissinger (ex USSecretary of State under Nixon) who immigrated to the US as a teenager whoretained a thick German accent, while his younger brother spoke like a native. JosephConrad’s accent was so strong, his speech was difficult to understand.

In a study done among Korean-born and Chinese-born immigrant students at IllinoisUniversity (Newport 1990), it was shown that immigrants who came to USA at agesof 3-7 performed identically in a test identifying sentences that containedgrammatical errors) to native students; those who arrived at ages 8-15 didincreasingly badly the later that they arrived. Those who arrived between ages of 17-39 did worst of all, doing increasingly badly according to age. (reported in pinker1994: 291)

42

Indeed there is an apparent paradox at the heart of any idea of evolution applied to aconcept such as language as it can be seen as at once benefiting the individual butalso dependent on other members of the group sharing the same faculty to at least asimilar level. Whereas collectively early language-users as a group may have gainedan advantage in being able to communicate and cooperate, thereby being able tooutwit and best physically better equipped animal competitors and predators,individual members of this humanoid group who were more proficient linguisticallythan their peers would have gained no immediate advantage from being so, at leastin early stages.

This is because being able to communicate more effectively and with greatersophistication is of little use if the rest of the group is unable to understand one. Bycontrast, a male golden eagle, for example, who has better eyesight than his peerswill have a clear individual competitive advantage that does not depend on theirrecognition or appreciation of his superior facilities. A more evolved eyesight willenable him to spot prey from longer distances and greater heights and with greaterprecision than other males will mean that he will be able to catch and eat more preyand thus grow ever healthier, fitter and physically stronger, which in turn will allowhim to establish a territory and fight off other males, find a mate, impress her with hisaerobatic courtship display and therefore reproduce.

By contrast, on the face it the advantages of having higher linguistic ability to an

43

individual are less obvious: if an early humanoid was born by some random geneticmutation (which is precisely how evolution would seem to work) with advancedlinguistic powers, these would not in themselves allow him or her to catch more preyor find more food or to be more attractive to potential mates. Indeed, the rest of thecommunity may be completely unaware of their superior linguistic powers in thesame way that parents of say a mathematical genius may be unaware of their child’stalent if they themselves are not numerate enough to recognise it: as Arthur ConanDoyle said in a Sherlock Holmes story (Valley of Fear, 1915): “mediocrity knowsnothing higher than itself”.

Two obvious explanations for why language may have benefited individuals presentthemselves and each regards the two key aspects of language namely that it is aninternal, cognitive process allowing certain kinds of thoughts to be conveyed ortranslated into a code that could then be converted into vocalisations and thentransmitted to others. The other aspect of language is the external one, that is how itserves not just to somehow convert thoughts into words inside a mind (Only certainkinds of thoughts because, not every that can be thought can be expressed bylanguage: language is by no means a perfect or complete system of communicationas philosophers such as Wittgenstein have stressed, as indeed our mental capacity forsight is fit for purpose but as an optical instrument flawed in many respects.), but alsoto communicate these words (and hopefully by extension the original thoughts) intothe minds of other people, allowing interlocutors to share ideas, knowledge andintentions with each other. The two are dependent on each other: without theinternal system, the external system could not exist but without the external system,there would be no need for an internal one in the first place.

As regards the internal system, while the advantages of being more proficientlinguistically than the rest of the group may not be immediately apparent, being lessproficient would certainly be a disadvantage as one would be unable to participatefully in the collective endeavours or one may fail to heed warnings (“don’t tease thesaber-tooth” to use the biologist George Williams’s example -- Cited by Pinker (1994:289)) and have one’s genes eventually weeded out through so-called naturalselection. There may have been other advantages related to the still obscure linksbetween thought and language, such as the fact that the possession of advancedlinguistic skills, regardless of whether these enabled the subject to communicatemore efficiently with the rest of the community, may nonetheless have had abeneficial effect on the subject’s capacity to indulge in rational and analyticalthought.

While the idea that language determines thought, often expressed in terms of withthe so-called Whorfian hypothesis or of linguistic relativity, finds very few proponentsamong linguistics or cognitive scientists today (see Pinker 1994, 1997, McWhorter

43

2014), most would agree that language can nonetheless affect thought in varioussubtle ways (see Deutscher 2010 and the influence in particular of the language ofspace – how it is used to describe the orientation of the world around speakers andevents). Certainly language, though principally serving to communicate with others,may also be used for what literary scholars call interior monologues that may allowpeople to “think through” solutions to various problems in a more extended anddetailed manner than might otherwise be possible (see for example Vygotsky (1986)on self-talk in children). Language also makes it easier to think in terms of extendedmetaphors which may likewise assist in problem solving and in the conceptualisationof ideas and phenomenon in the first place (see Lakoff and Johnson 2000).

As regards the external aspect of language that as a means of communication amongindividuals. In this light, language must have been a particular type of evolution whichdepended not only on advantage to the individual but also advantage to the group.Though evolution normally works primarily at the level of benefits to the individual,this does not exclude that a group may not play a role in the selection of traits whichincrease an individual’s ability to mate and have offspring, mainly by the fact that agroup may collectively come to value certain characteristics and thus privilege thoseamong them who display them. One example is the elaborate tail on a male peacock.Biologists believe that this not does not per se offer any evolutionary advantage tothe subject but it has come to be attractive to peahens presumably because by itsostentatious nature it signals clearer that the subject is so fit and healthy as to beable to spare valuable resources on what is essentially a display of pure opulence.

It is fair to presume that even early on in its evolution the language faculty came tobe highly valued by all members of a group that the mere possession of perceivablyadvanced linguistic proficiency would make the subject attractive to the rest of thecommunity and this increase his or her chances of finding a mate and reproducing, asArthur Conan Doyle went on to explain in the quote cited above: “mediocrity knowsnothing higher than itself, but talent instantly recognizes genius”. Were this the case,one might say that it constituted also the first green shoots of a culture as it wouldrepresent an awareness that there exists some social-conditioned activity thatconcerns the community as a whole, and whose raison d’être lies within that samecommunity, not just a single member of it. Indeed, it could be argued that therelationship between language and culture evolution-wise is circular and reciprocal,because the creation of culture has altered the “fitness landscape” that determineswhich exact hereditary traits give an individual advantage in terms of survival (seeKirby, 2011).

43

Language Interrupted, McWhorter (1987)

It also remains to seen whether “complicated” language really are or were spoken theway that traditional, prescriptive, grammars describe them. For example, the Latin ofCicero is notoriously precise and sophisticated yet most traces of everyday language(letters, graffiti) from Roman times indicates that ordinary people used a considerablysimpler “vulgar Latin”. Typically speakers of notoriously complicated language take apride in the fact that they are difficult to learn and speak, and revealingly peopleoften complain about dropping standards in the young and “contamination” fromoutside languages. It is thus possible that the much vaulted complexity is indeed away to exclude outsiders, and the often prescribed complexities are shibboleths.

Weber (1997): “the Chinese and to a lesser extent the Japanese, actually take theopposite attitude [to the French]. They consider their civilisations so manifestlysuperior that pressing their language on foreigners was really doing them too muchhonour. They also tend to think their languages far too complex to be mastered byclumsy strangers.”

English, Spanish and Chinese (not coincidentally the major world languages) arelanguages that have been adopted by diverse groups as second languages at keystages in their history (see for example so-called vulgar Latin in the Roman Empire).The fact that adults probably had to learn them as a second language may have lead

44

to simplification. For English this was the Danes in the Danelaw and perhaps theAncient Britons (Welsh) as well. The fact that complicated languages tend to bespoken by smaller more isolated communities and larger more widespreadlanguages are simpler supports this view

44

45

Learning is associated with traditional views of language as nurture; acquisition of language as nature.

This point will be developped further later on but it is good to get it clarified as soonas possible.

46

Bowen, C. (1998). Ages and Stages Summary: Language Development 0-5 years.

47

48

Although not all scholars agree with Chomsky, his demonstration that language isinnate has effectively closed the nature / nurture debate. Many would still attack UGas he describes it but are forced to find some kind of replacement for it.

In SLA research some scholars, allied to the nurture tradition, have argued that theUG provides an adequate explanation only for first language acquisition. Chomskyoriginally termed Language Acquisition Device (LAD) and this has led some to arguethat the device is activated just to let new-born babies acquire their first language(s).revealingly, Chomsky dropped the term in favour of UG. from an evolutionary point ofview, there is a clear logic to a time restraint on certain capacities (like also the abilityto reproduce). Our ancestors would probably have only ever needed to acquire thelanguages of the communities that they were born into and perhaps also those ofcontact groups and these would all have been encountered early in life (hence thefairly widespread phenomenon of bi-lingualism). Babies in communities where twolanguages are used become bi-lingual precisely because the LAD functions on bothlanguages at once, but this can only happen if both are first languages and neithertakes precedence.

49

According to this view, acquisition is a largely natural process, with instruction and teaching having little direct effect. UG remains relevant at all stages of life but failure in SLA is to be put down to factors which interfere with the natural UG.

50

As UF changes over time, being strongest in the years 0 to puberty, learners may needsome explicit information and instruction. There are also other outside factors thatmay counteract the UG.

51

Obviously L2 learners acquire the L2 at widely different rates; not all acheive anythinglike native-speaker competence, and, of course while it is perfectly normal to onlypartially acquire an L2, by definition, an L1 is never, in normal circumatsances, onlypartially acquired.

Newborns’ minds are free from “emotional clutter” and much more impressionableand UG can get easy access, with few distractions.

Even taking into account that it would seem that there is a CRITICAL PERIOD forinfluence of UG (0-12 years), it is also fair to assume that, as learners get older, theirminds are receptive to influences form various quarters and the UG’s influence isdiluted.

53

54

Order items.

External = curriculum, instruction, culture and status, motivation (also internal!), access to native speakers

The answers below are taken from a web-site. Some are questionable, however theyprovide an interesting starting point for discussion.

Internal factors.Age: Second language acquisition is influenced by the age of the learner. Children,who already have solid literacy skills in their own language, seem to be in the bestposition to acquire a new language efficiently. Motivated, older learners can be verysuccessful too, but usually struggle to achieve native-speaker-equivalentpronunciation and intonation.

Personality: Introverted or anxious learners usually make slower progress,particularly in the development of oral skills. They are less likely to take advantage ofopportunities to speak, or to seek out such opportunities. More outgoing studentswill not worry about the inevitability of making mistakes. They will take risks, andthus will give themselves much more practice.

55

Motivation: Intrinsic motivation has been found to correlate strongly witheducational achievement. Clearly, students who enjoy language learning and takepride in their progress will do better than those who don't.

Extrinsic motivation is also a significant factor. ESL students, for example, who needto learn English in order to take a place at an American university or to communicatewith a new English boy/girlfriend are likely to make greater efforts and thus greaterprogress.

Experiences: Learners who have acquired general knowledge and experience are in astronger position to develop a new language than those who haven't. The student,for example, who has already lived in 3 different countries and been exposed tovarious languages and cultures has a stronger base for learning a further languagethan the student who hasn't had such experiences.

Cognition: In general, it seems that students with greater cognitive abilities will makethe faster progress. Some linguists believe that there is a specific, innate languagelearning ability that is stronger in some students than in others.

Native language: Students who are learning a second language which is from thesame language family as their first language have, in general, a much easier task thanthose who aren't. So, for example, a Dutch child will learn English more quickly than aJapanese child.

External factors

External factors are those that characterize the particular language learning situation.

Curriculum: For ESL students in particular it is important that the totality of theireducational experience is appropriate for their needs. Language learning is less likelyto place if students are fully submersed into the mainstream program without anyextra assistance or, conversely, not allowed to be part of the mainstream until theyhave reached a certain level of language proficiency.

Instruction: Clearly, some language teachers are better than others at providingappropriate and effective learning experiences for the students in their classrooms.These students will make faster progress. The same applies to mainstream teachersin second language situations. The science teacher, for example, who is aware thatshe too is responsible for the students' English language development, and makescertain accommodations, will contribute to their linguistic development.

Culture and status: There is some evidence that students in situations where their

55

own culture has a lower status than that of the culture in which they are learning thelanguage make slower progress.

Motivation: Students who are given continuing, appropriate encouragement to learnby their teachers and parents will generally fare better than those who aren't. Forexample, students from families that place little importance on language learning arelikely to progress less quickly.

Access to native speakers: The opportunity to interact with native speakers bothwithin and outside of the classroom is a significant advantage. Native speakers arelinguistic models and can provide appropriate feedback. Clearly, second-languagelearners who have no extensive access to native speakers are likely to make slowerprogress, particularly in the oral/aural aspects of language acquisition.

http://esl.fis.edu/teachers/support/factors.htm

55

56

CALP (Cummins 1984) vs. BICS (Basic interactive communication skills).

57

58