北语英语语言学

rossy0622

是内部资料。

rossy0622

外国语言学与应用语言学专业备考纲要
外国语言学与应用语言学分四个科目的考试，其中包括：普通语言学，基础英语，二外，和政治理论。整个复习过程中需要特别注意的是语言学和基础英语，因为这两个科目各占150分，下面专门就普通语言学讲解一下题型及复习技巧。
语言学的参考书目只有一本，就是北大出版社出版的胡壮麟的《语言学教程》
第三版，考察的重点是关于语言本体的内容，即前五章导论，语音，词汇，句法，语义的内容，这一部分内容在考试中占到大于三分之二的分数，所以十分重要，后面几章中老师详细讲的只有，第七章社会语言学，第八章语用学，第十一章的语言学与文学，第十二章的流派，其他的不考。所以在复习过程中一定要把重点放在前五章语言主体的部分，概念及例子都要完全理解，要反复的排查自己不理解的地方，及时得寻求解决，否则这可能成为你最后考试中的致命伤，前五章既有概念填空题，又有语料分析题，更有定义和问答题的考察。切记一定要重点掌握，拿下了前五章，你的语言学复习就成功了一大半。后面几章看看老师的讲义还有期末考试的提纲就可以了。先不要急于背诵，首先先把课本从头到尾粗读一遍，了解大概讲了些什么内容，标出自己第一遍看不懂的地方，第二遍细读，就是一个知识点一个知识点得击破，着重理解第一遍没有看懂的地方，用不同的符号标出你对知识点的理解和掌握程度。第二遍结束以后，你已经对语言学的内容有了一个比较详细的了解，但是还没有达到说到一个知识点就能把它的英文定义，以及与它相联系的例子都说出来的水平，这证明你的复习还是远远不够的，到最后所要达到的水平是你对每一章节的内容了如指掌，能够在把书合上的时候，条理清晰得把它的内容表述出来。这样就达到了以不变应万变的效果。
第一遍和第二遍学习应该在八月份之前完成，当结束完了第二遍学习以后，就不用专门拿出时间来全书翻看了，毕竟考研并不是只考语言学，还有其他科目的学习，而且并不比语言学简单容易多少。相对而言，北语普通语言学考的都是基础的知识，不会考高深的没有学过的东西，所以大家不用太担心，只要把基础打好就完全没有问题，第三轮的复习就可以再复习其他科目的空闲时间把语言学里忘记的或是不会的地方反复去看，反复去记，直到学会记住为止。熟能生巧，少量多次原则在这轮复习里面非常奏效。也许你会觉得自己在这一段时间没有专门拿出时间来看语言学，但是其实你的知识点比以前掌握的更加深入了。不要着急，这个工作一般是在九月份和十月份之间完成，查漏补缺。力求每个知识点都理解到位。
接下来第四轮就是形成自己的体系，迅速得把书再看一遍，差不多在头脑中形成自己的记忆轮廓，不要急于在这个时候背诵，关键是形成一条线，把知识都串联起来。
在最后一个月的时候，就可以按照讲义来开始背诵了，这个过程有点痛苦，但是一定要坚持，坚持到底就是希望，可能对于有的同学来说语言学很枯燥，背了忘，忘了又背，没关系，多背几遍自然就可以了。考前一个星期每天都要背。
下面介绍一下语言学考试的题型。
1.填空题 ：50分 主要是知识点的理解和运用，大部分是语料分析，比如说给几个单词让分析哪部份是粘着词素，哪部份是自由词素。root和stem的区别，以及给出一个句子让分析里面词语的类别，如open class or closed class, functional word or lexical word ，以及划分句子成分的树形图，还有合作原则的各个次则。
2.填空题：50分  全都是概念，每个空2分，一共25个空，基本上都是课本和讲义上重点讲过的内容，没有太偏的题。当然也不会出讲义上的原话，而是把句法一变，不改变原来的意思，让你填写这个概念，只要理解得到位了，这个完全没有问题。
3.名词解释：一共有五到六个definition，做这样的题的时候只给出定义就可以了，如果学有余力的同学可以把例子也写上去，这样比较保险。以09年研究生入学考试为例：
Duality: is meant the property of having two levels of structures, such that units of the primary level are composed of elements of the secondary level and each of the two levels has its own principles of organization .we call sounds secondary units as opposed to primary units as words, since the secondary are meaningless and the primary unit have distinct and identifiable meaning.
Componential analysis: the meaning of a word is not an unanalysable whole. It may be seen as a complex of different semantic features. There are semantic units smaller than the meaning of a word. i.e. semantic components.
Paradigmatic relations are relations between a unit and other units that can replace it in a given sequence. Paradigmatically related units form systems.
Allophone
Root: the base form of a word that cannot be further analyzed without destroying its meaning.（The part of the word that is left when all the affixed are removed.）
4.问答题：两个问答题，09年出的题目是：
1.Leech’s seven types of meaning? 答这个题的时候，要把这七条记清楚，并最好附上例子，例子见后面的讲义。
1.        G.Leech recognizes 7 types for meaning in his semantics
1.        Conceptual meaning: logical, cognitive, or denotative content
2.        Connotative meaning: what is communicated by virtue of what language refers to.
3.        Social meaning: what is communicated of the social circumstances of language use.
4.        Affective meaning: what is communicated of the feelings and attitudes of the speaker/writer.
5.        Reflected meaning: what is communicated through association with another sense of the same expression.
6.        Collocative meaning: what is communicated through association with words which tend to occur in the environment of another word.
7.        Thematic meaning: what is communicated by the way in which the message is organized in terms of order and emphasis.
2.        The difference between subject-predicate analysis and theme-rheme analysis.  此题分别答一下两个的特点就可以了。
     In subject-predicate analysis, the subject is defined by the following properties:
1.        Word order: subject ordinarily precedes the verb in statement.
2.        Pro-forms: the first and third person pronouns in English appear in special form when the pronoun is a subject. This form is not used when the pronoun occurs in other position.
3.        Agreement with the verb: In the simple present tense, an –s is added to the verb when a third person subject is singular.
4.        Content question: if the subject is replaced by a question word (who/ what), the rest of the sentence remains the unchanged. But when any other element of the sentence is replaced by a question word, an auxiliary verb must appear before the subject.
5.        Tag question: the pronoun in a tag question normally corresponds to the subject of the sentence.
In theme-rheme analysis, Theme is the starting point of an utterance and Rheme is the nucleus or the core of the utterance. The analysis of a sentence in terms of theme and rheme is now known as the functional sentence perspective because this patterning is determined by the functional approach of the speaker.
二外日语，参考书目是新编标准日本语初级上下册，及中级上册，中级上册看旧版的就可以了。因为老师出题不可能从新版上出题。复习过程中一定把每课的单词和语法熟背，反复得背，另外需要注意的一点事课后的翻译题一定要背，而且要准确得背，因为翻译部分都是在这里出题的。09年的题目比较简单，出了初级上下册的题目，没有出中级的题，但是今年一定要三本都复习，以防万一。
题型：
1.汉字注假名（10个 有“海，歌，体，旅行，”）
2.假名写汉字（10个  有“老师，世界，电影，”）
3.写出外来语的意思（10个）
4.选择题，选助词
5.选择题，选实词
6.翻译   先是中翻日  然后是日翻中
7.一篇阅读理解，后面五个选择题。
基础英语，没有参考书目，按照专八的复习就可以了，完全可以应对，而且可以一举两得。推荐的书目有星火英语的标准阅读100篇，还有写作指南及翻译指南。不要用本科用的那种高级英语及综合教程。
题型
1.20个单词语法题
2.5篇阅读理解
3.一篇中翻英
4.一篇英翻中
5.作文
法语题型
1.        语法题：单项选择形式
2.        划线提问
3.        阅读
4.        作文 （两个题目，任选一写）：题目很简单，09年的两个题目为Ma famille & Ma vie universaire

rossy0622

Chapter 2 Speech Sounds
I. The three branches of phonetics
Articulatory phonetics: the study of sound production
Acoustic phonetics: the study of sound transmission between interlocutors
Auditory phonetics: the study of sound perception
II. Speech organs
Lungs肺
trachea (wind pipe)气管
vocal folds (cords)声带: glottis声门: apart (voiceless: /p/), closed together (voiced: /b/), totally closed (glottal stop: /?/
tongue tip舌尖
tongue blade舌叶
tongue front舌前
tongue back舌后
tongue root舌跟
epiglottis会厌
hard palate硬颚
soft palate (velum)软颚
uvula小舌
teeth 牙
teeth ridge (alveolar ridge) 齿龈
lips (labium)唇
nose鼻
larynx喉
pharynx咽
vocal tract声道
III. Segments, divergence, and phonetic transcriptions
Segments: smallest components of speech: bit -- /b/, /i/, /t/
Divergence: no one-to-one correspondent between pronunciation and spelling: ou – enough, house, through, though, etc.
Phonetic transcription: international phonetic alphabet
IV. Consonants
Consonants: when there is an obstruction of the air stream in the production of a sound
1) Manners of articulation:
Stop 爆破音: oral stop: /g/, nasal stop: /m/
Fricative摩擦音: /s/, /z/
(median or central) approximant央通音: /r/, /j/
Lateral approximant边通音: /l/
Affricate破擦音: /t∫/
2) Places of articulation
Bilabial唇音: /m/
Labiodental唇齿音: /f/
Dental齿音: /θ/
Alveolar齿龈音: /t/
Post-alveolar后齿龈音: /∫/
Palatal硬颚音: /j/
Velar软颚音: /k/
Glottal声门音: /h/
        Bilabial        Labiodental        dental        alveolar        Palatoalveolar        palatal        velar        glottal
nasal        m                        n                        ŋ       
ploasive        p b                         t d                        k ɡ       
fricative                f v        θ        s z        ∫ ʒ                        h
affricate                                        t∫ dʒ                       
Central approximant        (w)                        r                j        w       
Lateral approximant                                l                               
3) Voicing
4) Nasal vs oral
5) Lateral vs central
V. Vowels
Cardinal vowels: the reference points for the description and classification of vowels.
i                                             u

   e                                          o

       Є                                      ɔ

           a                                  ɑ
The part of the tongue that is raised: front /i:/, central /ə/, back /α/
The height of the tongue: high, mid, low; closed /i:/, half closed /e/, half open /ε/, open /a/
The degree of lip rounding: rounded /u:/, unrounded /i:/
Monophthong vs diphthong or pure vowels vs glidings: /a/, /au/
Long vs short vowels or tense vs lax vowels: /i:/, /I/
Front           central            back
High         i: (beet)                           u: (boot)
              I (bit)                           ʊ (put)
               e (bait)                         o (boat)
Mid              Є (bet)      ə ə: (worker)     ɔ: (bought)
⋀ (but)      ɒ (hot, Br.)
                  æ (bat)                    ɒ (hot, Am.)
low                 a (buy)                 ɑ: (farm)
VI. Coarticulation and phonetic transcription
Coarticulation协同发音: in speech a sound may become more like its neighbouring sound.
Anticipatory coarticulation逆化协同发音: when a sound is influenced by the following sound, e.g. impossible
Perseverative coarticulation接续性协同发音: when a sound is influenced by the preceding sound, e.g. play
Broad and narrow transcriptions 宽式和紧式音标
Diacritics: showing the minute differences between variations of the same sound, e.g. h for aspirated sounds, ~ for nasalized sounds, 0 for devoiced sounds. Broad transcriptions do not make use of these diacritics, while narrow transcriptions do.
VII. Phonological analysis/phonemic analysis
Phonemes音位: distinctive speech sounds
minimal pairs最小对立体: pairs of words that differ in only one sound
contrastive distribution对立分布: the two different sounds in a minimal pair are in contrastive distribution
Allophones音位变体: variants of the same phoneme
complementary distribution互补分布: those sounds that never occur in the same environment are in complementary distribution
phonetic similarity语音相似性: allophones of the same phoneme must be phonetically similar
pattern congruity模式一致性: when assigning a sound to one phoneme rather than another, we must take the sound pattern of the language into consideration
Phonological processes
Assimilation同化: when a sound take on some or all the characteristics of a neighbouring sound
regressive assimilation逆同化: when a sound is influenced by the following sound
progressive assimilation顺同化: when a sound is influenced by the preceding sound
Phonological processes: processes in which a sound undergoes a change in certain environments or contexts and we can write phonological rules to represent these changes: e.g.
Voiced sound → voiceless / voiceless _________
A voiced sound is transformed into a corresponding voiceless sound when it occurs after a voiceless sound: e.g. play
→ is transformed into; / specifies the environment in which the change occurs; _________ indicates the position of the target sound.
VIII. Distinctive features
Distinctive features: a phoneme can be further analysed into a set of features and the distinctive features are phonological; binary; articulatorily and acoustically based features
IX. Suprasegmental phonology超切分音系学concerned with those aspects of sound features that involve more than single sound segments
Suprasegmental phonemes超切分音位: stress, pitch, intonation
Syllables and syllabic structures: typically, a syllable consists of onset (consonant preceding the rhyme) and rhyme which consists of nucleus or peak (vowel or syllabic consonant) and coda (consonant following peak). (((C) (C) (C) V ((((C) (C) (C) (C)
Stress: degree of force used in the production of a syllable
word stress: distinctive, e.g. PERfect, perfect; REcord, record
sentence stress: normally, content words are stressed in sentences, while grammatical words are unstressed, but in principle, sentence stress can fall on any word or syllable
Pitch: different rates of the vibration of the vocal cords produce what is known in acoustics terms as different frequencies and in articulatory terms as different pithes. Pitch variations are called tones: tone languages and non-tone languages
Intonation: when pitch, stress and length variations are tied to the sentences rather than to the word, they are collectively known as intonation.
1) intonation can indicate different sentence types: yes-no interrogatives vs other types of sentences.
2) intonation can impose different sentence structures on sentences.
3) intonation can bring different part of the sentence into prominence.

rossy0622

Chapter 3 Lexicon
I. What is word
Three senses of word
1）        the physically definable unit: between pauses in speech and between blanks in writing
but “can’t”, “doesn’t”, “for ever”
2）        the common factor underlying a set of forms: work, works, worked, working
but “if”, “and”, “the”, “very”
3）a grammatical unit: sentence, clause, word group / phrase, word, morpheme
   but “nation”, “fast”
Identification of word
1)        stability: the constituent parts of a word have little potential for rearrangement
Playboy: *boyplay; John loves Mary: Mary loves John
2)        relative uninterruptibility: new elements and pauses are not to be inserted into a word
nationalization: *nationinteralization, *nation alization
(even) John doesn’t love Mary.
3)        a minimum free form: the minimal unit that can constitute an utterance by itself
-        Is Jane coming this evening? – Possibly.
-        What is missing in a sentence such as ‘Dog is barking’? – A.
Lexeme: basic abstract units of the lexicon on the level of language which may be realized in different grammatical forms such as the lexeme “work” in work, works, worked, working. A lexeme may also be a part of another lexeme, e.g. worker, workbook, workday. Idioms are also considered lexemes, e. g. the works of God: nature. Lexemes are the units which are conventionally listed in dictionaries as separate entries.
II. Classification of words
Variable vs invariable words
Variable words are words which have different grammatical forms.
Invariable words are words which do not have different grammatical forms.
Grammatical vs lexical words
Grammatical words are words which express grammatical meanings, e.g. pronouns, prepositions, conjunctions, etc. function words
Lexical words are words which have lexical content, e.g. nouns, verbs, etc. content words
Closed-class vs open-class words
Closed-class words are words whose membership is fixed or limited, e.g. conjunctions, pronouns, prepositions, etc. But even though prepositions are closed class words, they have been increasing in number slowly.
Open-class words are words whose membership is infinite or unlimited, e.g. nouns, verbs, etc.
Word classes: categories of words classified according to their grammatical, semantic, phonological properties or on the basis of formal similarities in terms of inflections and distribution. Nouns, verbs, adjectives, adverbs, pronouns, prepositions, conjunctions, interjections, articles
New word classes identified
Particles: the infinitive marker ‘to’ and the negative marker ‘not’ and the subordinate element in phrasal verbs, e.g. ‘at’ in look at and ‘out’ in knock out.
Auxiliaries: the traditional auxiliary and modal verbs such as ‘do’ and ‘can’
Pro-forms: There are other pro-forms in English in addition to pronouns.
pro-adjectives: Your pen is red. So is mine.
Pro-verbs: He knows English better than I do.
Pro-adverbs: He runs very fast and so do I.
pro-locative: Jane’s hiding there, behind the door.
Determiners: words preceding the head noun and determining the kind of reference the noun phrase has, e.g. the (definite) student, a (indefinite) student, some (partitive) students, all (universal) students.
Pre-diterminers: all, both, half, double, twice, one-third, etc.
Central determiners: (articles) the, a, (demonstrative pronouns) this, that, these, those, (indefinite pronouns) every, each, some, any, no, either, neither, (possessive pronouns) my, our, your, his, her, its, their, etc.
Post-determiners: (cardinal numerals) one, two, etc. (ordinal numerals) first, second, etc. (general ordinals) next, last, past, other, additional, etc. (quantifiers) many, several, much, little, a lot of, a great deal of, etc.
Determiners follow the order pre-determiner + central determiners + post determiners, e.g. all the students, half a year, etc.
The members of each subclass are usually exclusive of each other. But ordinal numerals and general ordinals may occur before cardinal numerals, e.g. the first two days, the past three weeks, etc.
III. Formation of words
Morpheme and morphology
Morphemes: minimal units of meaning
Types of morphemes:
free morphemes: can stand alone as words, e.g. dog, map, nation
bound morphemes: have to appear with at least another morpheme, e.g. international, pre cede, etc.
roots: that part of the word left when all the affixes are removed, e.g. internationalism, antidisestablishmentarianism
free roots: roots that can be used as words, e.g. hopeful, interpersonal
bound roots: roots that have to be appear with affixes, e.g. precede, receive, submit, retain, recur
some roots in English have both free and bound variants, e.g. sleepy, slept; childlike, children; goes, went
affixes: formative elements that have to be attached to at least another morpheme
prefixes: paragraph, miniskirt, unemployed, incorrect
suffixes: national, socialist, physics
infixes: foot / feet, man / men, in Cambodian /sepolah/ (field), /segepolah/ (fields)
stems: a morpheme or combination of morphemes to which an affix can be attached, e.g. national, internationalism
inflectional affixes and derivational affixes
1)        inflectional affixes are productive across an entire category, e.g. –s to all regular plural count nouns, but derivational affixes are not, e.g. production, connection, *maktion, increastion.
2)        inflectional affixes often add only a grammatical meaning to the stem, but derivational affixes often change the lexical content, e.g. maps, worker.
3)        Inflectional affixes do not change the word class of the stem, whereas derivational affixes may or may not change the word class of the stem, maps, smoker, incorrect.
4)        Inflectional affixes are often conditioned by non-semantic linguistic factors, e.g. John loves Mary. whereas derivational affixes are more often based on simple meaning distinctions, correct vs correctness.
5)        In English, inflectional affixes are normally suffixes, whereas derivational affixes can be prefixes as well.
6)        In English inflectional affixes are very small in number, whereas derivational affixes are much larger in number.
Inflection and word formation
Inflection: the manifestation of grammatical relationships through the addition of inflectional affixes, such as number, person, finiteness, aspect and case, which do not change the grammatical class of the stems to which they are attached.
Word formation: the process of word variations signaling lexical relationships. It can be further classified into the compositional type (compounding) and the derivational type (derivation).
Compound: words that consist of more than one free morpheme or the way to join two or more separate words to produce a single form.
Compounds can be completely united: playboy, hyphenated: mother-in-law, separated: high frequency.
There are noun compounds, e.g. daybreak, haircut, verb compounds, e.g. brainwash, lip-read, adjective compounds, e.g. man-eating, heartfelt, prepositional compounds, e.g. into, throughout. Compounds can be classified into endocentric （向心的） and exocentric （离心的） compounds. In endocentric nominal and adjectival compounds, the head is derived from a verb. Nominal: self-control, pain-killer, core-meaning; adjectival: eye-entertaining, bullet-resistant, machine-washable. In exocentric nominal and adjectival compounds, the first word is derived from the verb. Nominal: playboy, scarecrow, breakthrough, get-together; adjectival: take-home, runaway, drive-ins.
Derivation: showing the relationships between roots and affixes and changing or not changing the word class of the original words, e.g. unconscious, booklet, disobey, lengthen, foolish.
Phonology and morphology
1)        morpheme and phoneme
A single phoneme may represent a morpheme, but they are not identical, e.g. /z/: goes, boys, boy’s, is.
2) morphemic structure and phonological structure
Morphemes may also be represented by morphemic structures other than a single phoneme, e.g. love ly (monosyllabic), tobacco (polysyllabic). The syllabic structure of a word and its morphemic structure do not necessarily correspond.
3) allomorph
Some morphemes have a single form in all contexts, e.g. dog, cat, etc. other morphemes may have different shapes or phonetic forms, e.g. {z}: maps, dogs, watches, oxen, teeth, criteria, craft; {in}: incorrect, impossible, irregular, illegal.
4)        morphophonology or morphophonemics: the study of the relationship between phonology and morphonology.
(1)        phonologically conditioned
assimilation: maps, dogs, incorrect, impossible
dissimilation: peregrinus (Latin) → pilgrim, marbre (French) → marble
(2)        morphologically conditioned
three requirements:
a.        All the allomorphs should have common meaning, e.g. maps, dogs, watches.
b.        All the allomorphs should be in complementary distribution, e.g. maps, dogs, watches.
c.        Allomorphs that share the common meaning should be in parallel formation, e.g.
ox → oxen, cow → cows
IV. lexical changes
New words (neologisms, coinages) are created in the following ways:
Invention: Kodak, coke, nylon, fax, etc.
Compounding: moonwalk, chairperson, etc.
Derivation: semiconductor, supersonic
Blending: transfer + resister → transistor, smoke +fog → smog
        tele-printer +exchange → telex, modulator + demodulation → modem
Fusion ( a special type of blendings): spaddle ← spank（拍）+ paddle（打）, riffle ← ripple（波纹）+ ruffle（波纹）, rampacious ← rampageous（暴跳的） + rapacious（掠夺的）
Abbreviation: Clippings. Cutting the final part: advertisement → ad, mathematics → math; cutting the initial part: aeroplane → plane omnibus → bus; cutting both the initial and final parts: influenza → flu, refrigerator → fridge
Acronym: the initial letters of the words in a phrase or idiom or the name of an organization: EEC ← European Economic Community, CIA ← Central Intelligence Agency, Aids ← acquired immune deficiency syndrome, Radar ← radio detecting and ranging
Backformation: a shorter word is derived by deleting an imagined affix from a longer form: televise ← television, edit ← editor, diagnose ← diagnosis
Borrowings: from many different languages, especially Greek (electricity, atom), Latin (tumor, alibi), French (table, pork), Spanish (armada, ranch) etc.
Loanwords: both form and meaning are borrowed with only a light adaptation, e.g. coupon (French), sputnik (Russian), kung-fu (Chinese), Judo (Japanese).
Loanblends: part of the form is native and part is borrowed, but the meaning is fully borrowed, e.g. coconut (Spanish), Chinatown (Chinese),
loan shifts: the meaning is borrowed, but the form is native, e.g. bridge (a card game from Italian), artificial satellite (from Russian)
loan translation: each morpheme or word is translated from the equivalent morpheme or word in another language, e.g. free verse ← verse libre (Latin), black humour ← humour noir.
V. Phonological change
Loss: the disappearance of the sound as a phoneme in the phonological system, e.g. /x/ (the voiceless velar fricative) was lost between the times of Chaucer and Shakespeare: hit → it, niht → night. Sound loss may also occur in utterances at the expense of some unstressed vowels: temperature, postscript, the pen and pencil
Addition: Sounds may be added to the original sound sequence, e.g. rascal → rapscallion, lier → liaison
Metathesis: the alternation in the sequence of sounds, e.g. brid → bird, middel → middle, lytel → little.
Assimilation: the change of a sound as a result of the influence of an adjacent sound (contact or contiguous assimilation), e.g. immobile (n → m), support (b → p). Sometimes assimilation may occur between two sounds that are not too far apart (distant or non-contiguous assimilation), e.g. discussing shortly (s → ∫), confound it (∂ → au).
VI. Semantic change
Broadening: to extend or elevate the meaning of a word from its original specific sense to a relatively general one, e.g. offend (to strike against → to create or excite anger), bird (young bird → any kind of bird), companion (a person with whom you share bread → a person who accompanies you), etc.
Narrowing: the original meaning of a word is narrowed or restricted to a specific sense, e.g. girl (young person of either sex → young woman), deer (beast → a particular kind of animal), queen (wife → wife of a king or a female king), etc.
Meaning shift: the departure of a word from its original domain as a result of its metaphorical usage, e.g. bead (prayer → prayer bead → small, ball-shaped piece of glass, metal or wood), etc.
Class shift: conversion: the process in which a word changes from one word class to another, e.g. must (have to → an indispensable item), word (a language unit → to express in words), etc.

rossy0622

-Chapter 5 Meaning
Meanings of ‘meaning’
Leech’s seven types of meaning:
Conceptual meaning（central and basic）: logical, cognitive, denotative content, e.g. woman: a female human being, book: a number of printed pages bound together with a cover.
Associative meaning (peripheral):
Connotative meaning: What is communicated by virtue of what language refers to. Cross-individual differences, e.g. woman: long hair wearing, dress and skirt wearing, having maternal instinct, subject to instinct, prone to shed tears, physically weaker, inconstant, sociable and gregarious, emotional, compassionate, sensitive, tender and gentle, capable of speech, experienced in cookery, hard-working. Cross-cultural differences: individualism: (English) a doctrine claiming that the rights of individuals should be put over and above the rights of society, (Chinese) selfishness; book: (English) containing ideas for people to think about, to discuss, to make comments, (Chinese) containing knowledge for people to learn. There are synonyms differing in connotation: politician and statesman, farmer and peasant
Social meaning (stylistic meaning): What is communicated of the social circumstances of language use, e.g. torch and flash light, department for rent and flat to let; answer and reply, room and chamber; fire, flame and conflagration, weak, feeble and fragile; domicile, residence, abode and home, steed, horse and nag.
Affective meaning: what is communicated of the feelings and attitudes of the speaker / writer, e.g. you’re a liar. I hate you for that.
He had flashes of silence that made his conversation perfectly delightful.
For she was beautiful – her beauty made
The bright world dim, and everything beside
Seemed like the fleeting image of a shade.
“Generally speaking,” said Miss Murdstone, “I don’t like boys. How d’ye do, boy?” under these encouraging circumstances, I replied that I was very well, and wished that she was the same, with such indifferent grace that Miss Murdstone disposed of me in two words, -- “Wants manner！”
Reflected meaning: What is communicated through association with another sense of the same expression, e.g. The Comforter and The Holy Ghost, the morning star and the evening star; intercourse, ejaculation, erection and cook.
Collocative meaning: What is communicated through association with words which tend to occur in the environment of another word, e.g.
Pretty: girl, woman, flower, garden, colour, village
Handsome: boy, man, car, vessel, overcoat, airliner, typewriter
Thematic meaning: what is communicated by the way in which the message is organized in terms of order and emphasis, e.g.
Mr. Micawber has talent but doesn’t have capital.
Talent Mr. Micawber has, capital Mr. Micawber has not
The referential theory: a theory of meaning which relates the meaning of a word to the thing it refers to, or stands for.
Problems: 1) not every word has a reference, e.g. but, on, the. 2) a word is not related to a particular object in the world, e.g. book does not refer to a particular book. Its name, author, publisher, color, thickness and price do not matter when the word is used. The word refers to something abstract in language users’ mind, which is usually known as concept.
The semantic triangle (associated with Ogden and Richards):
                  concept

word   ------------------  thing
Sense refers to the abstract properties of an entity, while reference refers to concrete entities having these properties. In other words, Leech’s conceptual meaning has two sides: sense and reference. Every word has a sense, but not every word has a reference.
Sense relations
Synonymy: sameness relation. But total synonymy is rare. Synonyms may differ in style (Little Tom (bought, purchased) a toy bear.) and in connotation (At first I was poor, then I became needy, later I was underprivileged. Now I’m disadvantaged. I still don’t have a cent to my name, but I sure have a great vocabulary). There are dialectal differences as well, e.g. biscuit（英：饼干，美：软饼）, corn（英：谷物，美：玉米）, jumper（英：毛衣，美：套衫）, truck（英：铁路平板车，美：卡车）, store（英：货栈，美：商店）。
Antonymy: oppositeness relation.
Gradable antonymy: e.g. big - small, old - young, cold - hot, etc. these synonyms have three features: 1) they are gradable: they differ in terms of degree and can be modified by very and used in comparative and superlative degrees, 2) they are graded against different norms: a very small elephant is much bigger than a very big mouse, 3) usually the term for the higher degree serves as the core term (or unmarked term), long vs short.
Complementary antonymy: alive - dead, male - female, present - absent, odd - even. The assertion of one means the denial of the other and the denial of one means the assertion of the other. Complementary antonyms have three features, 1) they are similar to contradictory propositions in logic: they cannot be both true or false, e.g. This is a male cat. This is a female cat. In contrast, a pair of gradable antonyms can be compared to contrary propositions in logic: they cannot be both true, even though they can be both false, e.g. The coffee is hot. The coffee is cold. 2) the norm of this type is absolute, e.g. a male creature cannot be male in some situations while a female in other situations. 3) there is no cover term ( the unmarked term) for the two members of a pair.
Converse antonymy (relational opposites): the reversal of a relationship between two entities, e.g. buy – sell, lend – borrow, parent – child, host – guest, before – after. John borrowed something from Jane means the same as Jane lent something to John.
Hyponymy: meaning inclusiveness, e.g. furniture (superordinate) – desk, sofa, bed, chair, cupboard, etc. (hyponyms or co-hyponyms)
color
green   yellow      red
           scarlet   crimson    vermilion
living
plant         animal
      bird     fish   insect      animal
                         human        animal
                               tiger  wolf  elephant  monkey
Componential analysis: the meaning of a word is seen as a complex of semantic features or components, e.g.
boy: HUMAN, YOUNG, MALE;
girl: HUMAN, YOUNG, FEMALE;
man: HUMAN, ADULT, MALE;
woman: HUMAN, ADULT, FEMALE;
dog: CANINE, ADULT, MALE
bitch: CANINE, ADULT, FEMALE
puppy: CANINE, YOUNG
bull：BOVINE, ADULT, MALE
cow: BOVINE, ADULT, FEMALE
calf: BOVINE, YOUNG
If combined into binary features, then:
boy: HUMAN, -ADULT, MALE;
girl: HUMAN, -ADULT, -MALE;
man: HUMAN, ADULT, MALE;
woman: HUMAN, ADULT, -MALE.
Words that involve a relation between two entities can be shown:
Father: PARENT (x, y) & MALE (x) = x is the parent of y and x is male.
Mother: PARENT (x, y) & -MALE (x) = x is the parent of y and x is female.
Son: CHILD (x, y) & MALE (x) = x is the child of y and x is male.
Daughter: CHILD (x, y) & -MALE (x) = x is the child of y and x is female.
Verbs can also be analyzed in this way, e.g.
Take: CAUSE (x, (HAVE (x, y))) = x causes x to have y.
Give: CAUSE (x, (-HAVE (x, y))) = x causes x not to have y.
Die: BECOME (x, (-ALIVE (x))) = x becomes dead.
Kill: CAUSE (x, (BECOME (y, (-ALIVE (y)))) = x causes y to become dead
Sense relations may be better explained in terms of semantic components. Two words or expressions having the same semantic components will be synonyms, e.g. bachelor and unmarried man: HUMAN, ADULT, MALE, UNMARRIED. Two words having contrasting semantic components will be antonyms, e.g. man and woman, take and give, etc. Hyponyms have all the semantic components of their superordinates, e.g. boy and girl: HUMAN，-ADULT and child: HUMAN，-ADULT.
Problems with semantic componential analysis:
1)        Many words are polysemous. Consequently they have different sets of semantic components, e.g. man usually has the semantic component MALE, but it may also apply to both sexes, e.g. Man is mortal.
2)        Some semantic components are binary, e.g. MALE and FEMALE, ADULT and YOUNG. There is a clear-cut division line between MALE and FEMALE. But there is not such a clear-cut division line between ADULT and YOUNG. The division line between boy and man is fuzzy and that between girl and woman is even fuzzier.
3)        There may be words whose semantic features are difficult to ascertain. So whether it is possible to analyze all the lexical items in this way remains a question difficult to answer.
Semantic field theory (lexical field theory) is associated with primarily J. Trier. The following premises are fundamental to the lexical field theory.
1)        The meaning of an individual word is dependent upon the meaning of the rest of the words of the same lexical or conceptual field, e.g. evaluative words: good, excellent, exceptional, first-rate.
Military ranks
The USA                        China
General of the army（五星上将）   上将
General（上将）                 中将
Lieutenant general（中将）         少将
Major general（少将）            大校
Brigadier general（准将）          上校
Colonel（上校）                  中校
Lieutenant colonel（中校）         少校
Major（少校）                   上尉
Captain（上尉）                  中尉
First lieutenant（中尉）            少尉
Second lieutenant（少尉）          六级士官
Chief warrant officer（准尉）       五级士官
Master sergeant（军士长）         四级士官
Sergeant first class（上士）         三级士官
Sergeant（中士）                 二级士官
Corporal（下士）                 一级士官
Private first class（一等兵）        上等兵
Private（二等兵）                列兵
2)        An individual lexical field is constructed like a mosaic with no gaps; the whole set of all lexical fields of a language reflects a self-contained picture of reality.
3)        If a single word undergoes a change in meaning, then the structure of the lexical field changes.
Sentence meaning
Word meaning and sentence structure come together, e.g.
The boy chased the dog.
The dog chased the boy.
Predicate logic, also known as predicate calculus, studies the internal structure of simple propositions.
Proposition: what is expressed by a sentence which makes a statement. In predicate logic, a proposition has two parts: predicate and argument. For example,
John is a man. Predicate: man; argument: John. (one-place predicate)
John loves Mary: predicate: love; arguments: John, Mary. (two-place predicate)
John gave Mary a book. Predicate: give; arguments: John, Mary, book. (three-place predicate)
It rains. Predicate: rain. (zero-place predicate)
A sentence may contain several propositions, e.g.
John’ s friend Tony, who is a dentist, likes apples.
1)        John has a friend.
2)        The friend’s name is Tony.
3)        Tony is a dentist.
4) Tony likes apples.
A very important property of the proposition is that it has a truth value. It is either true or false, e.g.
The earth is round.
The earth is flat.
Propositional logic (propositional calculus or sentential calculus) is the study of the truth conditions for propositions: how the truth of a composite proposition (covering both compound and complex propositions) is determined by the truth value of its constituent （or component）propositions and the connections between them.
The truth value of a composite proposition is said to be the function of, or is determined by, the truth value of its component propositions and the logical connectives used in it. Negation (one-place connective): if a proposition p is true, then its negation ~p (orᆨp) is false, and if p is false, then ~p is true, e.g. John is married. John is not married. Conjunction (two-place connective): The logical connective is symbolized as & (or ∧). If (only when and as long as) both p and q are true, then p & q will be true, e.g. John was poor and (but) he was honest.  Disjunction (two-place connective): The logical connective is symbolized as ∨.Only when and as long as one of the component proposition is true (or both are true), the composite proposition p ∨ q will be true, e.g. Either he did not pass the driving test or I am a Dutchman. Implication (also known as conditional: two-place connective): the logical connective is symbolized as →. As long as the consequent proposition q is true, the composite proposition p → q will be true, e.g. If Ann has passed her driving test, her parents have bought her a Porsche. p → q is true (i) if Ann has passed her driving test and her parents have bought her a Porsche (p & q) , (ii) if she has not passed her driving test and (but) her parents have bought her a Porsche (~p & q), and (iii) Ann has not passed her driving test and her parents have not bought her a Porsche (~p & ~q). Equivalence (also known as biconditional or bilateral implication: two-place connective): the logical connective is symbolized as ≡ (or ↔). That is p ≡ q equals (p → q) & (q → p), e.g. Ralph is Philip’s father. → Philip is Ralph’s son and Philip is Ralph’s son. → Ralph is Philip’s father. The condition for the composite proposition p ≡ q to be true is that if and only if both component propositions are of the same truth value, whether true or false.
The truth table:
            and         or        if…then   if and only if…then
p     q        p & q        p ∨ q        p → q        p ≡ q
T     T            T            T            T            T
T     F            F            T            F            F
F     T            F            T            T            F
F     F            F            F            T            T
Entailment: a sense relation between sentences, e.g.
a. John killed Bill.
b.        Bill died
a.        I saw a boy.
b.        I saw a child
When a is true, b is necessarily true.
When b is false, a is false.
When a is false, b may be true or false.
When b is true, a may be true or false.
When these requirements are met, a entails b.
An integrated theory (associated with Katz, Fodor, Postal)
Katz and Fodor, 1963. “The Structure of a Semantic Theory”.
Katz and Postal, 1964. “An Integrated Theory of Language Description”.
The idea that the meaning of a sentence depends on the meaning of the constituent words and the way in which these words are combined is usually known as the principle of compositionality. The integrated theory tried to put this principle into practice: describing the internal structure of the semantic component.
According to them, the semantic theory consists of two parts: a dictionary and a set of projection rules. The dictionary provides the grammatical classification and semantic information of words. The former is more detailed than the traditional parts of speech or word classes, e.g. hit: a transitive verb {Vtr}, ball: a concrete noun {Nc}. {Vtr}, {Nc} and the like are termed as grammatical or syntactic markers. The latter is further classified into semantic markers and distinguishers. Semantic markers are more systematic and general, e.g. (Male), (Female), (Human), (Animal), etc. Distinguishers are more idiosyncratic or word specific, e.g. bachelor has the following distinguishers:
A [who has never married]
B [young knight serving under the standard of another knight]
C [who has the first or lowest academic degree]
D [young fur seal when without a mate during the breeding time]
The projection rules are responsible for combining the meanings of words together, e.g. The man hits the colorful ball: the meanings of the words in NP2 are combined first, then this NP2 is combined with the V, then the meanings of the words in NP1 are combined, then NP1 is combined with V + NP2. This effectively provides a solution to the integration of syntax and semantics.
In order to block the generation of sentences like Colorless green ideas sleep furiously, they also introduce some selection restrictions, e.g.
Colorful {Adj}
a. (color)[abounding in contrast or variety of bright colors]
<(Physical Object) or (Social Activity)>:
colorful ball, flowers, dress; colorful experiences, party, life
b. (evaluative)[having distinctive character, vividness, picturesqueness]
<(Aesthetic Object) or (Social Activity)>:
Colorful description, story, scenery
Ball {Nc}
a. (Social Activity)(Large)(Assembly)[for the purpose of social dancing]
b. (Physical Object)[having globular shape]
c. (Physical Object)[solid missile for projection by engine of war]
A projection rule will be in effect to combine the features of colorful and ball, resulting in the four readings of colorful ball:
a. (Social Activity)(Large)(Assembly)(color)[abounding in contrast or variety of bright colors] [for the purpose of social dancing]
b. (Physical Object) (color) [abounding in contrast or variety of bright colors] [having globular shape]
c. (Physical Object) (color) [abounding in contrast or variety of bright colors] [solid missile for projection by engine of war]
d. (Social Activity)(Large)(Assembly)(evaluative)[having distinctive character, vividness, picturesqueness] [for the purpose of social dancing]
The other two combinations of the second reading of colorful and the second or the third readings of ball are blocked by the selection restrictions.
In the end, the meanings of the whole sentence will be composed as shown below:
a.        [some contextually definite] (Physical Object) (Human) (Male) (Action) (Instancy) (Intensity) [collides with an impact] [some contextually definite] (Physical Object) (Color) [abounding in contrast or variety of bright colors] [having globular shape]
b. [some contextually definite] (Physical Object) (Human) (Male) (Action) (Instancy) (Intensity) [collides with an impact] [some contextually definite] (Physical Object) (Color) [abounding in contrast or variety of bright colors] [solid missile for projection by engine of war]
c. [some contextually definite] (Physical Object) (Human) (Male) (Action) (Instancy) (Intensity) [strike with a blow or missile] [some contextually definite] (Physical Object) (Color) [abounding in contrast or variety of bright colors] [having globular shape]
d. [some contextually definite] (Physical Object) (Human) (Male) (Action) (Instancy) (Intensity) [strike with a blow or missile] [some contextually definite] (Physical Object) (Color) [abounding in contrast or variety of bright colors] [solid missile for projection by engine of war]
Problems:
1)        The distinction between semantic marker and distinguisher is not very clear, e.g. having a globular shape can also be regarded as semantic marker since it is something general. Later, Katz dropped this distinction off.
2)        There are cases in which the collocation of words cannot be accounted for by grammatical markers, semantic markers or selection restrictions, e.g.
The girl gave her own dress away.
*The girl gave his own dress away.
My cousin is a male nurse.
My cousin is a female nurse.
3) This theory makes use of semantic markers which are elements of an artificial meta-language. To explain the meaning of the word man in terms of (Human), (Male), (Adult), one must go on to explain the meaning of theses semantic markers, otherwise they are meaningless.

The final examination:
1. Translation of terms in linguistics (20%)
  C – E (10%) 1 x 10
  E – C (10%) 1 x 10
2. Definition of terms (16%)
  4 x 4
3 – 10. Gap filling, analysis, etc. (50%)
11, 12. Essay questions (14%)

rossy0622

Chapter 6 Language Processing in Mind
I. introduction
Language processing in mind is studied in psycholinguistics concerned mainly with the storage, comprehension, production and acquisition of language.
Psycholinguistics attracts both linguists and psychologists. Linguists tend to favor descriptions of spontaneous speech as their main source of evidence, whereas psychologists mostly prefer experimental studies.
At its heart, psycholinguistic work consists of two questions:
1)        What knowledge of language is needed for us to use language?
Tacit knowledge: knowledge of how to perform various acts.
Explicit knowledge: knowledge of the processes or mechanisms used in these acts
Four broad areas of language knowledge: semantics, syntax, phonology, pragmatics
2)        What cognitive processes are involved in the ordinary use of language?
Garden path sentences:
The florist sent the flowers was very pleased.
Indirect requests:
Can you open the door, please?
Sentences with ambiguity:
The boy hit the girl with the boomerang.
Metaphors:
Some marriages are iceboxes.
The information processing system
Sensory stores
The sensory stores take in the variety of colors, tones, tastes, and smells that we experience each day and retain them, for a brief time, in a raw, unanalyzed form. It is assumed that we have one sensory store for each sensory system, although only the visual and auditory stores have been studied in any detail.
Working memory or short-term memory
The storage function: It is severely limited in size and can hold approximately 7 plus or minus 2 units of information for a limited time.
The processing function: it is related to the concept of processing capacity which refers to the total amount of cognitive resources we may devote to a task. When tasks are new or difficult, they require more processing capacity, thereby leaving less space available for the storage function.
Permanent memory or long-term memory
It is a repository of our knowledge of the world, including general knowledge and personal experiences. It holds all of the information we have retained from the past that is not currently active (not in working memory). These memories are used to interpret new experiences, and in turn the new events may be added to this storehouse of information.
Semantic memory: holding organized knowledge of words, concepts, symbols, and objects. It includes such broad classes of information as motor skills (typing, swimming, bicycling, etc.), general knowledge (grammar, mathematics, history, etc.), spatial knowledge (the typical layout of a house, etc.), social skills (how to begin and end a conversation, when and how to visit a friend, etc.).
Episodic memory: holding traces of events that are specific to a time and a place. This is the memory that we use to keep a record of our personal experiences. It thus varies from person to person and from time to time.
Their relevance for language processing
In comprehension, we may assume that as we hear a sentence the sounds are first stored briefly in the auditory sensory store. The sounds are held there for about 2 to 4 seconds, giving us more than enough time to recognize the auditory pattern. Pattern recognition occurs when information from one of the sensory stores is matched with information we retrieve from permanent memory. To recognize speech sounds we must identify some of the acoustic cues that are presented in the speech signal, such as the frequency of some of the sounds. At some point, after recognizing the sounds, we are able to organize them into syllables and eventually words, although it is not clear when and how this happens.
As noted, working memory can hold only about seven units of information. This could simply be seven words, but since many sentences are much longer than this, we need some way to deal immediately with more than seven words. One way we do this is to chunk the words into grammatical constituents such as noun and verb phrases, thereby reducing the storage burden to perhaps 2 or 3 constituents. The processing function of working memory is used to organize words into the constituents.
Permanent memory plays several roles. Semantic memory contains information on the speech sounds and words that we retrieve during pattern recognition. And while this process is going on, we are also building up an episodic memory representation of the ongoing discourse. That is, once we complete the processing of a given sentence, we might extract the gist of it and store that in episodic memory. Permanent memory would also be involved in finding antecedents for expressions such as pronouns; this would involve holding some information from previous sentences in memory long enough to establish coreferences.
Structure and process
It is generally agreed that the mind is likely to contain certain linguistic structures which are used in the course of various processes, such as comprehending and producing speech. Are these structures in language users’ mind the same as what is described in grammars? That is, are the grammatical structures described by linguists the same as the linguistic structures in language users’ mind? The grammatical structures described by linguists have psychological reality if they are the same as the linguistic structures in language users’ mind, because they are part of language processing. For example, in the standard theory of generative-transformational grammar the deep structures undergo transformations to become surface structures. If the transformations are really the mental processes when language users produce sentences these transformations have psychological reality.
Bottom-up and top-down processes
Bottom-up processing is defined as that which proceeds from the lowest level to the highest level of processing in such a way that all of the lower levels of processing operate without influence from the higher levels. That is the identification of phonemes is not affected by the lexical, syntactic, or discourse levels; the retrieval of words is not affected by syntactic or discourse levels, and so on.
A top-down processing model, in contrast, states that information at the higher levels may influence processing at the lower levels. For instance, a sentence context may affect the identification of words within that sentence. Speaking intuitively we may say that a top-down model of processing is one in which one’s expectations play a significant role.
Automatic and controlled processes
We may have a fixed processing capacity for handling information. Tasks that draw substantially from this limited pool of resources are called controlled tasks, and the processes involved in these tasks are controlled processes. Tasks that do not require substantial resources are called automatic tasks, and processes that do not require extensive capacity are referred to as automatic processes.
II. Language comprehension
Word recognition
One of the initial steps in understanding any message is the recognition of words. Word recognition is related to our knowledge of words, including their pronunciation (phonological knowledge), their internal structures (morphological knowledge), the word classes or parts of speech they belong to (syntactic knowledge), their definition or meaning (semantic knowledge). One common sense view of word recognition that receives a lot of support from experimental studies is the cohort theory. According to this theory, as soon as people hear speech, they start to narrow down the possible words that they may be hearing. If the first sound they hear is /s/, they eliminate all words beginning with other sounds; if the second sound is /p/, they eliminate many other possibilities. A word is recognized as soon as there is only one possibility left. That is, auditory word recognition begins with the formation of a group of words at the perception of the initial sound and proceeds sound by sound with the cohort of words decreasing as more sounds are perceived.
One of the most important factors that affect word recognition is frequency effect: how frequently the word is used in a given discourse or context. It explains the additional ease with which a word is accessed due to its more frequent usage in the language. For example, better is more frequently used than debtor.
People also recognized a word faster when they have just encountered it than when they have not recently encountered it. This is termed as recency effect. It describes the additional ease with which a word is recognized due to its repeated occurrence in the discourse or context. And it is possible to explain the frequency effect as the recency effect because frequent words are more likely to have been encountered than infrequent words.
Another factor that is involved in word recognition is context. People recognize a word more readily when the preceding words provide an appropriate context for it. For example, people recognize the word aorta faster in The heart surgeon carefully cut into the wall of the right aorta than in This is the aorta.
Lexical ambiguity
a.        After taking the right turn at the intersection … (ambiguous: correct vs rightward)
b.        After taking the left turn at the intersection … (unambiguous)
People take longer to understand a than b. This suggests that the two meanings of the word right are accessed. Hence one of the two main theories: all the meanings associated with the words are accessed. Other experiments suggest that in some situations only one meaning (the more frequent meaning or the meaning favored in a given context) is initially accessed. Hence the other of the two main theories: only one meaning is accessed initially.
Syntactic processing
Parsing
A first step in the process of understanding a sentence is to assign elements of its surface structure to linguistic categories. The result of parsing is an internal representation of the linguistic relationships within a sentence, usually in the form of a tree structure or phrase marker. For example,
                                     S
                                   
                          NP                 VP

                      Det      N         V         NP

                                               Det        N
The      actor     thanked  the      audience.
We may think of parsing as a form of problem solving or decision making in the sense that we making decisions about where to place incoming words into the phrase marker we are building. It has been suggested that we make these decisions immediately as we encounter a word, a principle known as immediacy principle. That is, when we first see or hear a word, we access its meaning from permanent memory, identify its likely referent, and fit it into the syntactic structure of the sentence. The alternative to immediate processing is to take a wait-and-see approach: to postpone interpreting a word or phrase until it is clearer where the sentence is going. However, there is considerable evidence for the immediacy principle. Although immediacy of processing reduces memory load, it may lead to errors. Considering the garden path sentence,
The florist sent the flowers was very pleased.
Sometimes it is not easy to determine which structures and meaning a sentence has. For example, there are two possible structures associated with the sentence The cop saw the spy with the binoculars. Some ambiguities are due to the ambiguous category of some of the words in the sentence. In the expression the desert trains, should desert be taken as the subject of the verb trains or is it a modifier of the noun trains?
The desert trains men to be hardy.
The desert trains seldom run on time.
Parsing strategies
If we are making decisions about where words fit into the syntactic structure of a sentence, on what are these decisions based? Much work has been done on the strategies we use in parsing. Strategies are thought of as approaches to parsing that work much of the time, although they are baldly foolproof. We will discuss two strategies that have gathered considerable empirical support. Late closure strategy One Parsing strategy is called the late closure strategy. This strategy states that, wherever possible, we prefer to attach new items to the current constituent. A primary motivation for this strategy is that it reduces the burden on working memory during parsing.
One example of late closure is the sentence below:
Tom said that Bill had taken the cleaning out yesterday.
Here the adverb yesterday may be attached to the main clause (Tom said …) or the subsequent subordinate clause (Bill had taken …) it has been argued that we tend to prefer the latter strategy.
Minimal attachment strategy A second strategy is referred to as the minimal attachment strategy, which states that we prefer attaching new items into the phrase marker being constructed using the fewest syntactic nodes consistent with the rules of the language. For example, a sentence fragment such as Ernie kissed Marcie and her sister . . . could be interpreted as either a noun phrase conjunction (that is, both Marcie and her sister were recipients of a kiss) or as the beginning of a new noun phrase. According to minimal attachment, we prefer the former interpretation.
Other evidence was found for the minimal attachment strategy. For example, consider the two sentences below:
The city council argued the mayor's position forcefully
The city council argued the mayor's position was correct.
The first sentence is consistent with minimal attachment in that the adverb forcefully is attached to the current constituent, the VP. In contrast, the second sentence is a complement construction that requires building a new constituent.
Semantics and sentence memory
Memory for Meaning versus surface form or wording
The key sentence:  a. Galileo, the great Italian scientist, sent a letter about it to him.
The test sentences: b. A letter about it was sent to him by Galileo, the great Italian scientist.
                c. Galileo, the great Italian scientist, sent him a letter about it.
                d. He sent a letter about it to Galileo, the great Italian scientist.
The results of the study were that if the test sentence was presented immediately after the key sentence, the subjects would recognize any change, syntactic or semantic. After as little as 60 syllables, however, subjects were likely to respond that they heard b or c. That is, after this short period of time they were not able to recognize syntactic changes that did not involve a change of meaning. For d the subjects’ performance was much more accurate and even after 120 syllables, they could still detect semantic changes with almost 100 percent accuracy. This seems to indicate that the syntactic details of linguistic material are not usually stored for a long time and that it is the semantic representation the subjects have available in their memory under normal circumstance.
Propositions and sentence memory
It appears that the exact linguistic form is not well retained and, moreover, additional, nonlinguistic information may play a major role in the retention process. Investigators have developed propositional models of sentence representation. All of the proposals assume that a sentence can be represented as a proposition consisting of two or more concepts and some form of relation between them. Thus, sentences (a), (b), (c) and (d), despite their superficial dissimilarities, all convey the same proposition.
(a) George hit Harry
Hit (George, Harry)
(b) Harry was hit by George.
(c) It was Harry who was hit by George
(d) The one who hit Harry was George.
More complex sentences convey more than one proposition. Sentence (e) could be represented as three separate propositions. Once again, these propositions may be realized linguistically in a very large number of ways
(e) George got into an argument with Harry, hit him, and then left the bar.
Initiated (George, Harry, argument)
Hit (George, Harry)
Left (George, bar)
A rough description of the way a propositional representation of a sentence might be set up during comprehension is as follows. When we first encounter a sentence, we extract its meaning and construct a proposition that represents this meaning. At the same time, the surface form of the sentence is being retained in working memory. Since the meaning is usually of greater interest, more processing resources are devoted to the meaning than to the surface form.
Inferences and Sentence Memory
It has been argued that we routinely draw inferences in the course of comprehending new events and that these inferences become incorporated into our memory representations of the event. With the passage of time, it becomes increasingly difficult to distinguish what was presented from what was inferred. In one study, people's comprehension and retention of the sentences below were examined:
John was trying to fix the birdhouse. He was looking for the nail when his father came out to watch him and to help him do the work.
John was using the hammer to fix the birdhouse when his father came out to watch him and to help him do the work.
The subjects believed that they heard the expression using the hammer which was not actually presented in the first sentence. The expression is what the subjects inferred from the first sentence and this inference became part of the subjects’ retention of the sentence.
Basic Processes in reading
Eye movement
A useful way of investigating some of the processes involved in reading is to study the eye movement of people reading. While we feel that our eyes move smoothly across a page of text, the reality is quite different. Our eyes actually make a series of rapid movements known as saccades, and between saccades there are fixation periods lasting approximately 250 milliseconds. A point towards the beginning of a word is usually fixated, and there is a distance of approximately eight letters or spaces between successive fixations. While most fixations typically move forwards in the text, around 10 or 15 % of them involve the eyes fixating an earlier part of the text than the previous fixation, known as regressions. Of particular importance, information is obtained from the text only during fixations and not at all during saccades.
The perceptual span
How much information is extracted from a single fixation. Various methods have been used to measure what is known as the PEHCEPTUAL SPAN, which is the range of letters from which useful information is extracted. Not surprisingly, the perceptual span varies depending on factors such as the size of the print, the complexity of the text, and so on. It is typically the case, however, that the perceptual span encompasses about three or four letters to the left of fixation and some fifteen letters to the right of fixation. What appears to be happening is that it is more valuable to look ahead in the text rather than to look backwards to words which have already been processed. The fact that the perceptual span covers almost 20 letters means that some of the letters included in it do not fall within the focal region of the eye, which is the area of high acuity. What information is extracted from the area lying outside the fovea? Fairly complex studies have revealed that meaning is not extracted, but information about the identity of the letters is obtained.
Discourse / text interpretations
When we are trying to understand a sentence, we often make use of information that that is not contained directly within the sentence itself. This is known as contextual information, and we can distinguish between two kinds of contextual: general and specific. GENERAL CONTEXT EFFECTS occur when our general knowledge about the world influences language comprehension. SPECIFIC CONTEXT EFFECTS involve information obtained from earlier parts of a discourse.
Schemata
It is believed that SCHEMATA, meaning packets of stored knowledge, play an important role in language processing. The features of schemata are as follows:
(1) Schemata can vary considerably in the information they contain, from the very simple to the very complex.
(2) Schemata are frequently organized hierarchically, for example, in addition to a rather general restaurant schema or script, we probably also have more specific restaurant schemata for different kinds of restaurant (e. g. fast-food places, up market French restaurants, and so on).
(3) Schemata operate on a top-down or conceptually driven way to facilitate interpretation of environmental stimuli.
Activation of appropriate schemata
We must activate the appropriate schemata to properly comprehend a story. The simplest case is the one in which we lack the appropriate schema. It has been demonstrated that the British college students had a very hard time understanding Eskimo folktales and tended to modify many of the details in their recall efforts, producing “a more coherent, concise, and undecorated tale.” It appears that comprehension and memory are poor when we do not have a schema that corresponds to the story that is unfolding, because it is nearly impossible to see the significance of the events being described.
In other instances, we may have an appropriate schema in memory but fail to activate it for one reason or another. A series of studies have convincingly demonstrated that comprehension and memory will be poor when the passage is written so obscurely that we cannot determine what might be the right schema, as in the following example:
With hocked gems financing him, our hero bravely defied all scornful laughter that tried to prevent his scheme. “Our eyes deceive,” he had said, “an egg not a table correctly typifies this unexplored planet.” Now three sturdy sisters sought proof, forging along sometimes through calm vastness, yet more often over turbulent peaks and valleys. Days became weeks as many doubters spread fearful rumors about the edge. At last from nowhere welcome winged creatures appeared signifying momentous success.
Reconstruction of schema-specific details
It has been argued that the activated schema serves as a retrieval plan, summoning up certain details rather than others by virtue of their centrality to the schema. Studies of comprehension with and without titles support this notion. For example, it was found that comprehenders who read a passage with one or two possible appropriate titles tended to emphasize different details in their recall. Thus, the perspective provided by the schema activated at the time of recoding seems to play an organizational role in our retrieval efforts.
Language production
More is known about language comprehension than language production.
Speech production
Speech production consists of four major stages: conceptualizing a thought to be expressed, formulating a linguistic plan, articulating the plan, and monitoring one’s speech.
Conceptualizing thoughts to be expressed
Very little can be said about the first stage. Basically, the questions here are, Where do ideas come from? And in what form do ideas exist before they are put into words? As to the latter question, psycholinguists and cognitive psychologists generally agree that some form of “mentalese” exists, that is, a representational system distinct from language. The notion is that thoughts take form in mentalese and are then translated into the linguistic form, but there is little agreement as to the properties of this pre-linguistic mental representation. The question of the origin of ideas may be even more intractable at this time, although some noteworthy efforts have been made to study the issue. Thus, we know that the first step occurs but are unable to say much about it.
Formulating linguistic plans
It has been suggested that the process of planning speech can be viewed as a series of stages, each devoted to one level of linguistic planning.
Stage Process
1    Identification of meaning – a meaning to be conveyed is generated
2    Selection of a syntactic structure – a syntactic outline of the sentence is constructed, with word slots specified.
3    Generation of intonation contour – the stress values of different word slots are assigned
4    Insertion of content words – appropriate nouns, verbs, and adjectives are retrieved from the lexicon and placed into word slots.
5   Formation of affixes and function words – function words (articles, conjunctions, prepositions, etc.) and suffixes are added.
6    Specification of phonetic segments – the sentence is expressed in terms of phonetic segments, according to phonological rules
The basic idea of this model is that we begin with the meaning that we wish to express and that subsequent levels of processing are devoted to specific and distinct aspects of the utterance. We set up a syntactic structure of the sentence, which specifies which words will receive major and minor stress and where the content words will fit in. Then the content words are added, followed by function words and affixes. Finally we identify the correct phonetic characteristics of the utterances, given this linguistic structure.
Implementing linguistic plans
Articulating
Once we have formulated our thoughts into a linguistic plan, this information must be sent from the brain to the speech muscles so that they can execute the required movements and produce the desired sounds. Fluent articulation of speech requires the coordinate use of a large number of speech muscles, which are distributed over three systems: the respiratory, the laryngeal, and the supra-laryngeal or vocal tract. These three systems of speech muscles control those bodily organs responsible for the production of speech sounds
Monitoring one’s speech
From time to time, we spontaneously interrupt our speech and correct ourselves. These corrections are referred to as self-repairs. They have a characteristic structure that consists of three parts. First we interrupt ourselves after we have detected an error in our speech. Second, we usually utter one of various editing expressions, including such terms as uh, sorry, I mean, and so forth. Finally we repair the utterance.
Corrections may occur within a word:
We can go straight on to the ye-, -- to the orange node.
They may also occur immediately after an error:
Straight on to green, -- to red.
They may be delayed by one or more words:
And from green left to pink – er from blue left to pink.
Although the major function of the editing expressions is to indicate that a correction follows, these editing expressions may have different connotations:
Bill hit him – hit Sam, that is. (to further specify a potentially ambiguous referent)
I am trying to lease, or rather, sublease, my apartment. (to substitute a word that is similar in meaning to the original word, but slightly closer to the speaker’s meaning)
I really like to – I mean, hate to – get up in the morning. (for true errors)
Three types of repairs have been identified:
Again left to the same blank crossing point – white crossing point. (instant repair)
And left to the purple crossing point – to the red crossing point. (anticipatory retracing)
From yellow down to brown – no – that’s red. (fresh start)

rossy0622

Chapter 6 Language Processing in Mind
I. introduction
Language processing in mind is studied in psycholinguistics concerned mainly with the storage, comprehension, production and acquisition of language.
Psycholinguistics attracts both linguists and psychologists. Linguists tend to favor descriptions of spontaneous speech as their main source of evidence, whereas psychologists mostly prefer experimental studies.
At its heart, psycholinguistic work consists of two questions:
1)        What knowledge of language is needed for us to use language?
Tacit knowledge: knowledge of how to perform various acts.
Explicit knowledge: knowledge of the processes or mechanisms used in these acts
Four broad areas of language knowledge: semantics, syntax, phonology, pragmatics
2)        What cognitive processes are involved in the ordinary use of language?
Garden path sentences:
The florist sent the flowers was very pleased.
Indirect requests:
Can you open the door, please?
Sentences with ambiguity:
The boy hit the girl with the boomerang.
Metaphors:
Some marriages are iceboxes.
The information processing system
Sensory stores
The sensory stores take in the variety of colors, tones, tastes, and smells that we experience each day and retain them, for a brief time, in a raw, unanalyzed form. It is assumed that we have one sensory store for each sensory system, although only the visual and auditory stores have been studied in any detail.
Working memory or short-term memory
The storage function: It is severely limited in size and can hold approximately 7 plus or minus 2 units of information for a limited time.
The processing function: it is related to the concept of processing capacity which refers to the total amount of cognitive resources we may devote to a task. When tasks are new or difficult, they require more processing capacity, thereby leaving less space available for the storage function.
Permanent memory or long-term memory
It is a repository of our knowledge of the world, including general knowledge and personal experiences. It holds all of the information we have retained from the past that is not currently active (not in working memory). These memories are used to interpret new experiences, and in turn the new events may be added to this storehouse of information.
Semantic memory: holding organized knowledge of words, concepts, symbols, and objects. It includes such broad classes of information as motor skills (typing, swimming, bicycling, etc.), general knowledge (grammar, mathematics, history, etc.), spatial knowledge (the typical layout of a house, etc.), social skills (how to begin and end a conversation, when and how to visit a friend, etc.).
Episodic memory: holding traces of events that are specific to a time and a place. This is the memory that we use to keep a record of our personal experiences. It thus varies from person to person and from time to time.
Their relevance for language processing
In comprehension, we may assume that as we hear a sentence the sounds are first stored briefly in the auditory sensory store. The sounds are held there for about 2 to 4 seconds, giving us more than enough time to recognize the auditory pattern. Pattern recognition occurs when information from one of the sensory stores is matched with information we retrieve from permanent memory. To recognize speech sounds we must identify some of the acoustic cues that are presented in the speech signal, such as the frequency of some of the sounds. At some point, after recognizing the sounds, we are able to organize them into syllables and eventually words, although it is not clear when and how this happens.
As noted, working memory can hold only about seven units of information. This could simply be seven words, but since many sentences are much longer than this, we need some way to deal immediately with more than seven words. One way we do this is to chunk the words into grammatical constituents such as noun and verb phrases, thereby reducing the storage burden to perhaps 2 or 3 constituents. The processing function of working memory is used to organize words into the constituents.
Permanent memory plays several roles. Semantic memory contains information on the speech sounds and words that we retrieve during pattern recognition. And while this process is going on, we are also building up an episodic memory representation of the ongoing discourse. That is, once we complete the processing of a given sentence, we might extract the gist of it and store that in episodic memory. Permanent memory would also be involved in finding antecedents for expressions such as pronouns; this would involve holding some information from previous sentences in memory long enough to establish coreferences.
Structure and process
It is generally agreed that the mind is likely to contain certain linguistic structures which are used in the course of various processes, such as comprehending and producing speech. Are these structures in language users’ mind the same as what is described in grammars? That is, are the grammatical structures described by linguists the same as the linguistic structures in language users’ mind? The grammatical structures described by linguists have psychological reality if they are the same as the linguistic structures in language users’ mind, because they are part of language processing. For example, in the standard theory of generative-transformational grammar the deep structures undergo transformations to become surface structures. If the transformations are really the mental processes when language users produce sentences these transformations have psychological reality.
Bottom-up and top-down processes
Bottom-up processing is defined as that which proceeds from the lowest level to the highest level of processing in such a way that all of the lower levels of processing operate without influence from the higher levels. That is the identification of phonemes is not affected by the lexical, syntactic, or discourse levels; the retrieval of words is not affected by syntactic or discourse levels, and so on.
A top-down processing model, in contrast, states that information at the higher levels may influence processing at the lower levels. For instance, a sentence context may affect the identification of words within that sentence. Speaking intuitively we may say that a top-down model of processing is one in which one’s expectations play a significant role.
Automatic and controlled processes
We may have a fixed processing capacity for handling information. Tasks that draw substantially from this limited pool of resources are called controlled tasks, and the processes involved in these tasks are controlled processes. Tasks that do not require substantial resources are called automatic tasks, and processes that do not require extensive capacity are referred to as automatic processes.
II. Language comprehension
Word recognition
One of the initial steps in understanding any message is the recognition of words. Word recognition is related to our knowledge of words, including their pronunciation (phonological knowledge), their internal structures (morphological knowledge), the word classes or parts of speech they belong to (syntactic knowledge), their definition or meaning (semantic knowledge). One common sense view of word recognition that receives a lot of support from experimental studies is the cohort theory. According to this theory, as soon as people hear speech, they start to narrow down the possible words that they may be hearing. If the first sound they hear is /s/, they eliminate all words beginning with other sounds; if the second sound is /p/, they eliminate many other possibilities. A word is recognized as soon as there is only one possibility left. That is, auditory word recognition begins with the formation of a group of words at the perception of the initial sound and proceeds sound by sound with the cohort of words decreasing as more sounds are perceived.
One of the most important factors that affect word recognition is frequency effect: how frequently the word is used in a given discourse or context. It explains the additional ease with which a word is accessed due to its more frequent usage in the language. For example, better is more frequently used than debtor.
People also recognized a word faster when they have just encountered it than when they have not recently encountered it. This is termed as recency effect. It describes the additional ease with which a word is recognized due to its repeated occurrence in the discourse or context. And it is possible to explain the frequency effect as the recency effect because frequent words are more likely to have been encountered than infrequent words.
Another factor that is involved in word recognition is context. People recognize a word more readily when the preceding words provide an appropriate context for it. For example, people recognize the word aorta faster in The heart surgeon carefully cut into the wall of the right aorta than in This is the aorta.
Lexical ambiguity
a.        After taking the right turn at the intersection … (ambiguous: correct vs rightward)
b.        After taking the left turn at the intersection … (unambiguous)
People take longer to understand a than b. This suggests that the two meanings of the word right are accessed. Hence one of the two main theories: all the meanings associated with the words are accessed. Other experiments suggest that in some situations only one meaning (the more frequent meaning or the meaning favored in a given context) is initially accessed. Hence the other of the two main theories: only one meaning is accessed initially.
Syntactic processing
Parsing
A first step in the process of understanding a sentence is to assign elements of its surface structure to linguistic categories. The result of parsing is an internal representation of the linguistic relationships within a sentence, usually in the form of a tree structure or phrase marker. For example,
                                     S
                                   
                          NP                 VP

                      Det      N         V         NP

                                               Det        N
The      actor     thanked  the      audience.
We may think of parsing as a form of problem solving or decision making in the sense that we making decisions about where to place incoming words into the phrase marker we are building. It has been suggested that we make these decisions immediately as we encounter a word, a principle known as immediacy principle. That is, when we first see or hear a word, we access its meaning from permanent memory, identify its likely referent, and fit it into the syntactic structure of the sentence. The alternative to immediate processing is to take a wait-and-see approach: to postpone interpreting a word or phrase until it is clearer where the sentence is going. However, there is considerable evidence for the immediacy principle. Although immediacy of processing reduces memory load, it may lead to errors. Considering the garden path sentence,
The florist sent the flowers was very pleased.
Sometimes it is not easy to determine which structures and meaning a sentence has. For example, there are two possible structures associated with the sentence The cop saw the spy with the binoculars. Some ambiguities are due to the ambiguous category of some of the words in the sentence. In the expression the desert trains, should desert be taken as the subject of the verb trains or is it a modifier of the noun trains?
The desert trains men to be hardy.
The desert trains seldom run on time.
Parsing strategies
If we are making decisions about where words fit into the syntactic structure of a sentence, on what are these decisions based? Much work has been done on the strategies we use in parsing. Strategies are thought of as approaches to parsing that work much of the time, although they are baldly foolproof. We will discuss two strategies that have gathered considerable empirical support. Late closure strategy One Parsing strategy is called the late closure strategy. This strategy states that, wherever possible, we prefer to attach new items to the current constituent. A primary motivation for this strategy is that it reduces the burden on working memory during parsing.
One example of late closure is the sentence below:
Tom said that Bill had taken the cleaning out yesterday.
Here the adverb yesterday may be attached to the main clause (Tom said …) or the subsequent subordinate clause (Bill had taken …) it has been argued that we tend to prefer the latter strategy.
Minimal attachment strategy A second strategy is referred to as the minimal attachment strategy, which states that we prefer attaching new items into the phrase marker being constructed using the fewest syntactic nodes consistent with the rules of the language. For example, a sentence fragment such as Ernie kissed Marcie and her sister . . . could be interpreted as either a noun phrase conjunction (that is, both Marcie and her sister were recipients of a kiss) or as the beginning of a new noun phrase. According to minimal attachment, we prefer the former interpretation.
Other evidence was found for the minimal attachment strategy. For example, consider the two sentences below:
The city council argued the mayor's position forcefully
The city council argued the mayor's position was correct.
The first sentence is consistent with minimal attachment in that the adverb forcefully is attached to the current constituent, the VP. In contrast, the second sentence is a complement construction that requires building a new constituent.
Semantics and sentence memory
Memory for Meaning versus surface form or wording
The key sentence:  a. Galileo, the great Italian scientist, sent a letter about it to him.
The test sentences: b. A letter about it was sent to him by Galileo, the great Italian scientist.
                c. Galileo, the great Italian scientist, sent him a letter about it.
                d. He sent a letter about it to Galileo, the great Italian scientist.
The results of the study were that if the test sentence was presented immediately after the key sentence, the subjects would recognize any change, syntactic or semantic. After as little as 60 syllables, however, subjects were likely to respond that they heard b or c. That is, after this short period of time they were not able to recognize syntactic changes that did not involve a change of meaning. For d the subjects’ performance was much more accurate and even after 120 syllables, they could still detect semantic changes with almost 100 percent accuracy. This seems to indicate that the syntactic details of linguistic material are not usually stored for a long time and that it is the semantic representation the subjects have available in their memory under normal circumstance.
Propositions and sentence memory
It appears that the exact linguistic form is not well retained and, moreover, additional, nonlinguistic information may play a major role in the retention process. Investigators have developed propositional models of sentence representation. All of the proposals assume that a sentence can be represented as a proposition consisting of two or more concepts and some form of relation between them. Thus, sentences (a), (b), (c) and (d), despite their superficial dissimilarities, all convey the same proposition.
(a) George hit Harry
Hit (George, Harry)
(b) Harry was hit by George.
(c) It was Harry who was hit by George
(d) The one who hit Harry was George.
More complex sentences convey more than one proposition. Sentence (e) could be represented as three separate propositions. Once again, these propositions may be realized linguistically in a very large number of ways
(e) George got into an argument with Harry, hit him, and then left the bar.
Initiated (George, Harry, argument)
Hit (George, Harry)
Left (George, bar)
A rough description of the way a propositional representation of a sentence might be set up during comprehension is as follows. When we first encounter a sentence, we extract its meaning and construct a proposition that represents this meaning. At the same time, the surface form of the sentence is being retained in working memory. Since the meaning is usually of greater interest, more processing resources are devoted to the meaning than to the surface form.
Inferences and Sentence Memory
It has been argued that we routinely draw inferences in the course of comprehending new events and that these inferences become incorporated into our memory representations of the event. With the passage of time, it becomes increasingly difficult to distinguish what was presented from what was inferred. In one study, people's comprehension and retention of the sentences below were examined:
John was trying to fix the birdhouse. He was looking for the nail when his father came out to watch him and to help him do the work.
John was using the hammer to fix the birdhouse when his father came out to watch him and to help him do the work.
The subjects believed that they heard the expression using the hammer which was not actually presented in the first sentence. The expression is what the subjects inferred from the first sentence and this inference became part of the subjects’ retention of the sentence.
Basic Processes in reading
Eye movement
A useful way of investigating some of the processes involved in reading is to study the eye movement of people reading. While we feel that our eyes move smoothly across a page of text, the reality is quite different. Our eyes actually make a series of rapid movements known as saccades, and between saccades there are fixation periods lasting approximately 250 milliseconds. A point towards the beginning of a word is usually fixated, and there is a distance of approximately eight letters or spaces between successive fixations. While most fixations typically move forwards in the text, around 10 or 15 % of them involve the eyes fixating an earlier part of the text than the previous fixation, known as regressions. Of particular importance, information is obtained from the text only during fixations and not at all during saccades.
The perceptual span
How much information is extracted from a single fixation. Various methods have been used to measure what is known as the PEHCEPTUAL SPAN, which is the range of letters from which useful information is extracted. Not surprisingly, the perceptual span varies depending on factors such as the size of the print, the complexity of the text, and so on. It is typically the case, however, that the perceptual span encompasses about three or four letters to the left of fixation and some fifteen letters to the right of fixation. What appears to be happening is that it is more valuable to look ahead in the text rather than to look backwards to words which have already been processed. The fact that the perceptual span covers almost 20 letters means that some of the letters included in it do not fall within the focal region of the eye, which is the area of high acuity. What information is extracted from the area lying outside the fovea? Fairly complex studies have revealed that meaning is not extracted, but information about the identity of the letters is obtained.
Discourse / text interpretations
When we are trying to understand a sentence, we often make use of information that that is not contained directly within the sentence itself. This is known as contextual information, and we can distinguish between two kinds of contextual: general and specific. GENERAL CONTEXT EFFECTS occur when our general knowledge about the world influences language comprehension. SPECIFIC CONTEXT EFFECTS involve information obtained from earlier parts of a discourse.
Schemata
It is believed that SCHEMATA, meaning packets of stored knowledge, play an important role in language processing. The features of schemata are as follows:
(1) Schemata can vary considerably in the information they contain, from the very simple to the very complex.
(2) Schemata are frequently organized hierarchically, for example, in addition to a rather general restaurant schema or script, we probably also have more specific restaurant schemata for different kinds of restaurant (e. g. fast-food places, up market French restaurants, and so on).
(3) Schemata operate on a top-down or conceptually driven way to facilitate interpretation of environmental stimuli.
Activation of appropriate schemata
We must activate the appropriate schemata to properly comprehend a story. The simplest case is the one in which we lack the appropriate schema. It has been demonstrated that the British college students had a very hard time understanding Eskimo folktales and tended to modify many of the details in their recall efforts, producing “a more coherent, concise, and undecorated tale.” It appears that comprehension and memory are poor when we do not have a schema that corresponds to the story that is unfolding, because it is nearly impossible to see the significance of the events being described.
In other instances, we may have an appropriate schema in memory but fail to activate it for one reason or another. A series of studies have convincingly demonstrated that comprehension and memory will be poor when the passage is written so obscurely that we cannot determine what might be the right schema, as in the following example:
With hocked gems financing him, our hero bravely defied all scornful laughter that tried to prevent his scheme. “Our eyes deceive,” he had said, “an egg not a table correctly typifies this unexplored planet.” Now three sturdy sisters sought proof, forging along sometimes through calm vastness, yet more often over turbulent peaks and valleys. Days became weeks as many doubters spread fearful rumors about the edge. At last from nowhere welcome winged creatures appeared signifying momentous success.
Reconstruction of schema-specific details
It has been argued that the activated schema serves as a retrieval plan, summoning up certain details rather than others by virtue of their centrality to the schema. Studies of comprehension with and without titles support this notion. For example, it was found that comprehenders who read a passage with one or two possible appropriate titles tended to emphasize different details in their recall. Thus, the perspective provided by the schema activated at the time of recoding seems to play an organizational role in our retrieval efforts.
Language production
More is known about language comprehension than language production.
Speech production
Speech production consists of four major stages: conceptualizing a thought to be expressed, formulating a linguistic plan, articulating the plan, and monitoring one’s speech.
Conceptualizing thoughts to be expressed
Very little can be said about the first stage. Basically, the questions here are, Where do ideas come from? And in what form do ideas exist before they are put into words? As to the latter question, psycholinguists and cognitive psychologists generally agree that some form of “mentalese” exists, that is, a representational system distinct from language. The notion is that thoughts take form in mentalese and are then translated into the linguistic form, but there is little agreement as to the properties of this pre-linguistic mental representation. The question of the origin of ideas may be even more intractable at this time, although some noteworthy efforts have been made to study the issue. Thus, we know that the first step occurs but are unable to say much about it.
Formulating linguistic plans
It has been suggested that the process of planning speech can be viewed as a series of stages, each devoted to one level of linguistic planning.
Stage Process
1    Identification of meaning – a meaning to be conveyed is generated
2    Selection of a syntactic structure – a syntactic outline of the sentence is constructed, with word slots specified.
3    Generation of intonation contour – the stress values of different word slots are assigned
4    Insertion of content words – appropriate nouns, verbs, and adjectives are retrieved from the lexicon and placed into word slots.
5   Formation of affixes and function words – function words (articles, conjunctions, prepositions, etc.) and suffixes are added.
6    Specification of phonetic segments – the sentence is expressed in terms of phonetic segments, according to phonological rules
The basic idea of this model is that we begin with the meaning that we wish to express and that subsequent levels of processing are devoted to specific and distinct aspects of the utterance. We set up a syntactic structure of the sentence, which specifies which words will receive major and minor stress and where the content words will fit in. Then the content words are added, followed by function words and affixes. Finally we identify the correct phonetic characteristics of the utterances, given this linguistic structure.
Implementing linguistic plans
Articulating
Once we have formulated our thoughts into a linguistic plan, this information must be sent from the brain to the speech muscles so that they can execute the required movements and produce the desired sounds. Fluent articulation of speech requires the coordinate use of a large number of speech muscles, which are distributed over three systems: the respiratory, the laryngeal, and the supra-laryngeal or vocal tract. These three systems of speech muscles control those bodily organs responsible for the production of speech sounds
Monitoring one’s speech
From time to time, we spontaneously interrupt our speech and correct ourselves. These corrections are referred to as self-repairs. They have a characteristic structure that consists of three parts. First we interrupt ourselves after we have detected an error in our speech. Second, we usually utter one of various editing expressions, including such terms as uh, sorry, I mean, and so forth. Finally we repair the utterance.
Corrections may occur within a word:
We can go straight on to the ye-, -- to the orange node.
They may also occur immediately after an error:
Straight on to green, -- to red.
They may be delayed by one or more words:
And from green left to pink – er from blue left to pink.
Although the major function of the editing expressions is to indicate that a correction follows, these editing expressions may have different connotations:
Bill hit him – hit Sam, that is. (to further specify a potentially ambiguous referent)
I am trying to lease, or rather, sublease, my apartment. (to substitute a word that is similar in meaning to the original word, but slightly closer to the speaker’s meaning)
I really like to – I mean, hate to – get up in the morning. (for true errors)
Three types of repairs have been identified:
Again left to the same blank crossing point – white crossing point. (instant repair)
And left to the purple crossing point – to the red crossing point. (anticipatory retracing)
From yellow down to brown – no – that’s red. (fresh start)

rossy0622

Chapter 6 Language Processing in Mind
I. introduction
Language processing in mind is studied in psycholinguistics concerned mainly with the storage, comprehension, production and acquisition of language.
Psycholinguistics attracts both linguists and psychologists. Linguists tend to favor descriptions of spontaneous speech as their main source of evidence, whereas psychologists mostly prefer experimental studies.
At its heart, psycholinguistic work consists of two questions:
1)        What knowledge of language is needed for us to use language?
Tacit knowledge: knowledge of how to perform various acts.
Explicit knowledge: knowledge of the processes or mechanisms used in these acts
Four broad areas of language knowledge: semantics, syntax, phonology, pragmatics
2)        What cognitive processes are involved in the ordinary use of language?
Garden path sentences:
The florist sent the flowers was very pleased.
Indirect requests:
Can you open the door, please?
Sentences with ambiguity:
The boy hit the girl with the boomerang.
Metaphors:
Some marriages are iceboxes.
The information processing system
Sensory stores
The sensory stores take in the variety of colors, tones, tastes, and smells that we experience each day and retain them, for a brief time, in a raw, unanalyzed form. It is assumed that we have one sensory store for each sensory system, although only the visual and auditory stores have been studied in any detail.
Working memory or short-term memory
The storage function: It is severely limited in size and can hold approximately 7 plus or minus 2 units of information for a limited time.
The processing function: it is related to the concept of processing capacity which refers to the total amount of cognitive resources we may devote to a task. When tasks are new or difficult, they require more processing capacity, thereby leaving less space available for the storage function.
Permanent memory or long-term memory
It is a repository of our knowledge of the world, including general knowledge and personal experiences. It holds all of the information we have retained from the past that is not currently active (not in working memory). These memories are used to interpret new experiences, and in turn the new events may be added to this storehouse of information.
Semantic memory: holding organized knowledge of words, concepts, symbols, and objects. It includes such broad classes of information as motor skills (typing, swimming, bicycling, etc.), general knowledge (grammar, mathematics, history, etc.), spatial knowledge (the typical layout of a house, etc.), social skills (how to begin and end a conversation, when and how to visit a friend, etc.).
Episodic memory: holding traces of events that are specific to a time and a place. This is the memory that we use to keep a record of our personal experiences. It thus varies from person to person and from time to time.
Their relevance for language processing
In comprehension, we may assume that as we hear a sentence the sounds are first stored briefly in the auditory sensory store. The sounds are held there for about 2 to 4 seconds, giving us more than enough time to recognize the auditory pattern. Pattern recognition occurs when information from one of the sensory stores is matched with information we retrieve from permanent memory. To recognize speech sounds we must identify some of the acoustic cues that are presented in the speech signal, such as the frequency of some of the sounds. At some point, after recognizing the sounds, we are able to organize them into syllables and eventually words, although it is not clear when and how this happens.
As noted, working memory can hold only about seven units of information. This could simply be seven words, but since many sentences are much longer than this, we need some way to deal immediately with more than seven words. One way we do this is to chunk the words into grammatical constituents such as noun and verb phrases, thereby reducing the storage burden to perhaps 2 or 3 constituents. The processing function of working memory is used to organize words into the constituents.
Permanent memory plays several roles. Semantic memory contains information on the speech sounds and words that we retrieve during pattern recognition. And while this process is going on, we are also building up an episodic memory representation of the ongoing discourse. That is, once we complete the processing of a given sentence, we might extract the gist of it and store that in episodic memory. Permanent memory would also be involved in finding antecedents for expressions such as pronouns; this would involve holding some information from previous sentences in memory long enough to establish coreferences.
Structure and process
It is generally agreed that the mind is likely to contain certain linguistic structures which are used in the course of various processes, such as comprehending and producing speech. Are these structures in language users’ mind the same as what is described in grammars? That is, are the grammatical structures described by linguists the same as the linguistic structures in language users’ mind? The grammatical structures described by linguists have psychological reality if they are the same as the linguistic structures in language users’ mind, because they are part of language processing. For example, in the standard theory of generative-transformational grammar the deep structures undergo transformations to become surface structures. If the transformations are really the mental processes when language users produce sentences these transformations have psychological reality.
Bottom-up and top-down processes
Bottom-up processing is defined as that which proceeds from the lowest level to the highest level of processing in such a way that all of the lower levels of processing operate without influence from the higher levels. That is the identification of phonemes is not affected by the lexical, syntactic, or discourse levels; the retrieval of words is not affected by syntactic or discourse levels, and so on.
A top-down processing model, in contrast, states that information at the higher levels may influence processing at the lower levels. For instance, a sentence context may affect the identification of words within that sentence. Speaking intuitively we may say that a top-down model of processing is one in which one’s expectations play a significant role.
Automatic and controlled processes
We may have a fixed processing capacity for handling information. Tasks that draw substantially from this limited pool of resources are called controlled tasks, and the processes involved in these tasks are controlled processes. Tasks that do not require substantial resources are called automatic tasks, and processes that do not require extensive capacity are referred to as automatic processes.
II. Language comprehension
Word recognition
One of the initial steps in understanding any message is the recognition of words. Word recognition is related to our knowledge of words, including their pronunciation (phonological knowledge), their internal structures (morphological knowledge), the word classes or parts of speech they belong to (syntactic knowledge), their definition or meaning (semantic knowledge). One common sense view of word recognition that receives a lot of support from experimental studies is the cohort theory. According to this theory, as soon as people hear speech, they start to narrow down the possible words that they may be hearing. If the first sound they hear is /s/, they eliminate all words beginning with other sounds; if the second sound is /p/, they eliminate many other possibilities. A word is recognized as soon as there is only one possibility left. That is, auditory word recognition begins with the formation of a group of words at the perception of the initial sound and proceeds sound by sound with the cohort of words decreasing as more sounds are perceived.
One of the most important factors that affect word recognition is frequency effect: how frequently the word is used in a given discourse or context. It explains the additional ease with which a word is accessed due to its more frequent usage in the language. For example, better is more frequently used than debtor.
People also recognized a word faster when they have just encountered it than when they have not recently encountered it. This is termed as recency effect. It describes the additional ease with which a word is recognized due to its repeated occurrence in the discourse or context. And it is possible to explain the frequency effect as the recency effect because frequent words are more likely to have been encountered than infrequent words.
Another factor that is involved in word recognition is context. People recognize a word more readily when the preceding words provide an appropriate context for it. For example, people recognize the word aorta faster in The heart surgeon carefully cut into the wall of the right aorta than in This is the aorta.
Lexical ambiguity
a.        After taking the right turn at the intersection … (ambiguous: correct vs rightward)
b.        After taking the left turn at the intersection … (unambiguous)
People take longer to understand a than b. This suggests that the two meanings of the word right are accessed. Hence one of the two main theories: all the meanings associated with the words are accessed. Other experiments suggest that in some situations only one meaning (the more frequent meaning or the meaning favored in a given context) is initially accessed. Hence the other of the two main theories: only one meaning is accessed initially.
Syntactic processing
Parsing
A first step in the process of understanding a sentence is to assign elements of its surface structure to linguistic categories. The result of parsing is an internal representation of the linguistic relationships within a sentence, usually in the form of a tree structure or phrase marker. For example,
                                     S
                                   
                          NP                 VP

                      Det      N         V         NP

                                               Det        N
The      actor     thanked  the      audience.
We may think of parsing as a form of problem solving or decision making in the sense that we making decisions about where to place incoming words into the phrase marker we are building. It has been suggested that we make these decisions immediately as we encounter a word, a principle known as immediacy principle. That is, when we first see or hear a word, we access its meaning from permanent memory, identify its likely referent, and fit it into the syntactic structure of the sentence. The alternative to immediate processing is to take a wait-and-see approach: to postpone interpreting a word or phrase until it is clearer where the sentence is going. However, there is considerable evidence for the immediacy principle. Although immediacy of processing reduces memory load, it may lead to errors. Considering the garden path sentence,
The florist sent the flowers was very pleased.
Sometimes it is not easy to determine which structures and meaning a sentence has. For example, there are two possible structures associated with the sentence The cop saw the spy with the binoculars. Some ambiguities are due to the ambiguous category of some of the words in the sentence. In the expression the desert trains, should desert be taken as the subject of the verb trains or is it a modifier of the noun trains?
The desert trains men to be hardy.
The desert trains seldom run on time.
Parsing strategies
If we are making decisions about where words fit into the syntactic structure of a sentence, on what are these decisions based? Much work has been done on the strategies we use in parsing. Strategies are thought of as approaches to parsing that work much of the time, although they are baldly foolproof. We will discuss two strategies that have gathered considerable empirical support. Late closure strategy One Parsing strategy is called the late closure strategy. This strategy states that, wherever possible, we prefer to attach new items to the current constituent. A primary motivation for this strategy is that it reduces the burden on working memory during parsing.
One example of late closure is the sentence below:
Tom said that Bill had taken the cleaning out yesterday.
Here the adverb yesterday may be attached to the main clause (Tom said …) or the subsequent subordinate clause (Bill had taken …) it has been argued that we tend to prefer the latter strategy.
Minimal attachment strategy A second strategy is referred to as the minimal attachment strategy, which states that we prefer attaching new items into the phrase marker being constructed using the fewest syntactic nodes consistent with the rules of the language. For example, a sentence fragment such as Ernie kissed Marcie and her sister . . . could be interpreted as either a noun phrase conjunction (that is, both Marcie and her sister were recipients of a kiss) or as the beginning of a new noun phrase. According to minimal attachment, we prefer the former interpretation.
Other evidence was found for the minimal attachment strategy. For example, consider the two sentences below:
The city council argued the mayor's position forcefully
The city council argued the mayor's position was correct.
The first sentence is consistent with minimal attachment in that the adverb forcefully is attached to the current constituent, the VP. In contrast, the second sentence is a complement construction that requires building a new constituent.
Semantics and sentence memory
Memory for Meaning versus surface form or wording
The key sentence:  a. Galileo, the great Italian scientist, sent a letter about it to him.
The test sentences: b. A letter about it was sent to him by Galileo, the great Italian scientist.
                c. Galileo, the great Italian scientist, sent him a letter about it.
                d. He sent a letter about it to Galileo, the great Italian scientist.
The results of the study were that if the test sentence was presented immediately after the key sentence, the subjects would recognize any change, syntactic or semantic. After as little as 60 syllables, however, subjects were likely to respond that they heard b or c. That is, after this short period of time they were not able to recognize syntactic changes that did not involve a change of meaning. For d the subjects’ performance was much more accurate and even after 120 syllables, they could still detect semantic changes with almost 100 percent accuracy. This seems to indicate that the syntactic details of linguistic material are not usually stored for a long time and that it is the semantic representation the subjects have available in their memory under normal circumstance.
Propositions and sentence memory
It appears that the exact linguistic form is not well retained and, moreover, additional, nonlinguistic information may play a major role in the retention process. Investigators have developed propositional models of sentence representation. All of the proposals assume that a sentence can be represented as a proposition consisting of two or more concepts and some form of relation between them. Thus, sentences (a), (b), (c) and (d), despite their superficial dissimilarities, all convey the same proposition.
(a) George hit Harry
Hit (George, Harry)
(b) Harry was hit by George.
(c) It was Harry who was hit by George
(d) The one who hit Harry was George.
More complex sentences convey more than one proposition. Sentence (e) could be represented as three separate propositions. Once again, these propositions may be realized linguistically in a very large number of ways
(e) George got into an argument with Harry, hit him, and then left the bar.
Initiated (George, Harry, argument)
Hit (George, Harry)
Left (George, bar)
A rough description of the way a propositional representation of a sentence might be set up during comprehension is as follows. When we first encounter a sentence, we extract its meaning and construct a proposition that represents this meaning. At the same time, the surface form of the sentence is being retained in working memory. Since the meaning is usually of greater interest, more processing resources are devoted to the meaning than to the surface form.
Inferences and Sentence Memory
It has been argued that we routinely draw inferences in the course of comprehending new events and that these inferences become incorporated into our memory representations of the event. With the passage of time, it becomes increasingly difficult to distinguish what was presented from what was inferred. In one study, people's comprehension and retention of the sentences below were examined:
John was trying to fix the birdhouse. He was looking for the nail when his father came out to watch him and to help him do the work.
John was using the hammer to fix the birdhouse when his father came out to watch him and to help him do the work.
The subjects believed that they heard the expression using the hammer which was not actually presented in the first sentence. The expression is what the subjects inferred from the first sentence and this inference became part of the subjects’ retention of the sentence.
Basic Processes in reading
Eye movement
A useful way of investigating some of the processes involved in reading is to study the eye movement of people reading. While we feel that our eyes move smoothly across a page of text, the reality is quite different. Our eyes actually make a series of rapid movements known as saccades, and between saccades there are fixation periods lasting approximately 250 milliseconds. A point towards the beginning of a word is usually fixated, and there is a distance of approximately eight letters or spaces between successive fixations. While most fixations typically move forwards in the text, around 10 or 15 % of them involve the eyes fixating an earlier part of the text than the previous fixation, known as regressions. Of particular importance, information is obtained from the text only during fixations and not at all during saccades.
The perceptual span
How much information is extracted from a single fixation. Various methods have been used to measure what is known as the PEHCEPTUAL SPAN, which is the range of letters from which useful information is extracted. Not surprisingly, the perceptual span varies depending on factors such as the size of the print, the complexity of the text, and so on. It is typically the case, however, that the perceptual span encompasses about three or four letters to the left of fixation and some fifteen letters to the right of fixation. What appears to be happening is that it is more valuable to look ahead in the text rather than to look backwards to words which have already been processed. The fact that the perceptual span covers almost 20 letters means that some of the letters included in it do not fall within the focal region of the eye, which is the area of high acuity. What information is extracted from the area lying outside the fovea? Fairly complex studies have revealed that meaning is not extracted, but information about the identity of the letters is obtained.
Discourse / text interpretations
When we are trying to understand a sentence, we often make use of information that that is not contained directly within the sentence itself. This is known as contextual information, and we can distinguish between two kinds of contextual: general and specific. GENERAL CONTEXT EFFECTS occur when our general knowledge about the world influences language comprehension. SPECIFIC CONTEXT EFFECTS involve information obtained from earlier parts of a discourse.
Schemata
It is believed that SCHEMATA, meaning packets of stored knowledge, play an important role in language processing. The features of schemata are as follows:
(1) Schemata can vary considerably in the information they contain, from the very simple to the very complex.
(2) Schemata are frequently organized hierarchically, for example, in addition to a rather general restaurant schema or script, we probably also have more specific restaurant schemata for different kinds of restaurant (e. g. fast-food places, up market French restaurants, and so on).
(3) Schemata operate on a top-down or conceptually driven way to facilitate interpretation of environmental stimuli.
Activation of appropriate schemata
We must activate the appropriate schemata to properly comprehend a story. The simplest case is the one in which we lack the appropriate schema. It has been demonstrated that the British college students had a very hard time understanding Eskimo folktales and tended to modify many of the details in their recall efforts, producing “a more coherent, concise, and undecorated tale.” It appears that comprehension and memory are poor when we do not have a schema that corresponds to the story that is unfolding, because it is nearly impossible to see the significance of the events being described.
In other instances, we may have an appropriate schema in memory but fail to activate it for one reason or another. A series of studies have convincingly demonstrated that comprehension and memory will be poor when the passage is written so obscurely that we cannot determine what might be the right schema, as in the following example:
With hocked gems financing him, our hero bravely defied all scornful laughter that tried to prevent his scheme. “Our eyes deceive,” he had said, “an egg not a table correctly typifies this unexplored planet.” Now three sturdy sisters sought proof, forging along sometimes through calm vastness, yet more often over turbulent peaks and valleys. Days became weeks as many doubters spread fearful rumors about the edge. At last from nowhere welcome winged creatures appeared signifying momentous success.
Reconstruction of schema-specific details
It has been argued that the activated schema serves as a retrieval plan, summoning up certain details rather than others by virtue of their centrality to the schema. Studies of comprehension with and without titles support this notion. For example, it was found that comprehenders who read a passage with one or two possible appropriate titles tended to emphasize different details in their recall. Thus, the perspective provided by the schema activated at the time of recoding seems to play an organizational role in our retrieval efforts.
Language production
More is known about language comprehension than language production.
Speech production
Speech production consists of four major stages: conceptualizing a thought to be expressed, formulating a linguistic plan, articulating the plan, and monitoring one’s speech.
Conceptualizing thoughts to be expressed
Very little can be said about the first stage. Basically, the questions here are, Where do ideas come from? And in what form do ideas exist before they are put into words? As to the latter question, psycholinguists and cognitive psychologists generally agree that some form of “mentalese” exists, that is, a representational system distinct from language. The notion is that thoughts take form in mentalese and are then translated into the linguistic form, but there is little agreement as to the properties of this pre-linguistic mental representation. The question of the origin of ideas may be even more intractable at this time, although some noteworthy efforts have been made to study the issue. Thus, we know that the first step occurs but are unable to say much about it.
Formulating linguistic plans
It has been suggested that the process of planning speech can be viewed as a series of stages, each devoted to one level of linguistic planning.
Stage Process
1    Identification of meaning – a meaning to be conveyed is generated
2    Selection of a syntactic structure – a syntactic outline of the sentence is constructed, with word slots specified.
3    Generation of intonation contour – the stress values of different word slots are assigned
4    Insertion of content words – appropriate nouns, verbs, and adjectives are retrieved from the lexicon and placed into word slots.
5   Formation of affixes and function words – function words (articles, conjunctions, prepositions, etc.) and suffixes are added.
6    Specification of phonetic segments – the sentence is expressed in terms of phonetic segments, according to phonological rules
The basic idea of this model is that we begin with the meaning that we wish to express and that subsequent levels of processing are devoted to specific and distinct aspects of the utterance. We set up a syntactic structure of the sentence, which specifies which words will receive major and minor stress and where the content words will fit in. Then the content words are added, followed by function words and affixes. Finally we identify the correct phonetic characteristics of the utterances, given this linguistic structure.
Implementing linguistic plans
Articulating
Once we have formulated our thoughts into a linguistic plan, this information must be sent from the brain to the speech muscles so that they can execute the required movements and produce the desired sounds. Fluent articulation of speech requires the coordinate use of a large number of speech muscles, which are distributed over three systems: the respiratory, the laryngeal, and the supra-laryngeal or vocal tract. These three systems of speech muscles control those bodily organs responsible for the production of speech sounds
Monitoring one’s speech
From time to time, we spontaneously interrupt our speech and correct ourselves. These corrections are referred to as self-repairs. They have a characteristic structure that consists of three parts. First we interrupt ourselves after we have detected an error in our speech. Second, we usually utter one of various editing expressions, including such terms as uh, sorry, I mean, and so forth. Finally we repair the utterance.
Corrections may occur within a word:
We can go straight on to the ye-, -- to the orange node.
They may also occur immediately after an error:
Straight on to green, -- to red.
They may be delayed by one or more words:
And from green left to pink – er from blue left to pink.
Although the major function of the editing expressions is to indicate that a correction follows, these editing expressions may have different connotations:
Bill hit him – hit Sam, that is. (to further specify a potentially ambiguous referent)
I am trying to lease, or rather, sublease, my apartment. (to substitute a word that is similar in meaning to the original word, but slightly closer to the speaker’s meaning)
I really like to – I mean, hate to – get up in the morning. (for true errors)
Three types of repairs have been identified:
Again left to the same blank crossing point – white crossing point. (instant repair)
And left to the purple crossing point – to the red crossing point. (anticipatory retracing)
From yellow down to brown – no – that’s red. (fresh start)

rossy0622

Chapter 8 language in use
“A policeman is coming.” : The meaning of the words + structural meaning
                      A description of something in reality
                      A warning
The second type of meaning is sometimes referred to as speaker’s meaning, utterance meaning or contextual meaning, and the discipline that concentrates on this type of meaning is called pragmatics = meaning - semantics.
Speech act theory
This is the first major theory in the study of language in use. It originated with the Oxford philosopher Austin and presented in How to Do Things with Words.
Performatives（言语行为句）and constatives（表述句）
Austin classifies sentences in two categories: performatives and constatives. Performatives do not describe things. They can not be said to be true or false. Uttering them is, or is a part of, doing an action, e.g.
I name this ship Queen Elizabeth.
I apologize.
I declare the meeting open.
I sentence you ten years of imprisonment。
Constatives are descriptive statements, capable of being analyzed in terms of truth-values, e.g.
I pour some liquid into the tube.
Felicity conditions（适宜条件）of the performatives:
A (i) There must be a relevant conventional procedure, and
  (ii) The relevant participants and circumstances must be appropriate.
B The procedure must be executed (i) correctly and (ii) completely.
C Very often, (i) the relevant people must have the requisite thoughts, feelings and intentions, and (ii) must follow it up with actions specified.
However, Austin soon realized that these conditions could only apply to some cases. For example, I apologize and I promise to come early tomorrow may be produced without a strict procedure. And, on the other hand, The present king of France is bald, is infelicitous in the same way as I bequeath my watch to my brother said by someone without a watch.
Then Austin explored the possibility of separating performatives from constatives on grammatical and lexical criteria. Typical performatives use:
the first person singular pronoun as the subject,
the indicative mood,
the simple present tense,
the active voice,
and performative verbs.
But there are counterexamples:
Pedestrians are warned to keep off the grass. (passive) (I warn you to keep off the grass.)
Turn right. (imperative) (I order you to turn right.)
You did it. (past) (I find you guilty.)
Thank you (imperative) (I thank you.)
And state can be used as a performative verb, e.g.
I state that I’m alone responsible.
It seems that the distinction between performatives and constatives can not be maintained. All sentences can be used to do things.
A theory of the illocutionary act
According to Austin, there are three senses in which saying something may be understood as doing something
It’s cold in here.
Locutionary act（表述性言语行为、言中行为）: the act of producing speech sounds, words or sentences.
Illocutionary act（施为性言语行为、言外行为）: the act of making known the speaker’s purpose or the intended meaning: asking or answering a question, giving some information or an assurance or a warning, pronouncing sentence, making a request or an appointment or a criticism, making an identification or giving a description, and many others.
Austin acknowledges that force or illocutionary force can be regarded as part of meaning, when the latter is used in a broad sense. But he thinks it is better to distinguish force from meaning, with the latter used in a narrow sense, or what we say the more constant, inherent side of meaning (studied in semantics). Thus interpreted, force may be said to be equivalent to speaker’s meaning, contextual meaning, or extra meaning (studied in pragmatics).
Perlocutionary act（言后行为）: the consequential effect of a locution upon the hearer. By saying something the speaker may change the opinion of the hearer, misleading him, surprising him, or inducing him to do something. Whether these effects are intended by the speaker, they can be regarded as part of the act performed by the speaker. This is what is called perlocutionary act.
The theory about illocutionary acts is the central issue in the speech act theory. One can even say that the speech act theory is in fact a theory of the illocutionary act. In this general theory, the original performatives are only special types in which the illocutionary force is made explicit by the performative verb.
The theory of conversational implicature（会话含义）
Associated with another Oxford philosopher H.B. Grice and presented mainly in his Logic and Conversation.
Grice noticed that in daily conversations people do not usually say things directly but tend to imply them. For example, when A said to B about C, “Oh he’s quite well I think. He likes his colleagues, and he hasn’t been to prison yet.” A certainly implied something though he did not say it explicitly. Grice coined the term implicature （avoid the term implication used in semantics）to refer to this type of implied or suggested meaning. And he investigated the ways in which people manage to convey implicature.
According to Grice, in daily conversations, people are cooperative. They often recognize a common purpose or a set of purposes or at least a mutually agreed direction for the conversation to develop. That is, they follow a cooperative principle or CP for short. The cooperative principle has four categories of maxims:
Quantity
1.        make your contribution as informative as is required (for the current purpose of the exchange)
2.        do not make your contribution more informative than is required.
Quality
1.        do not say what you believe to be false.
2.        do not say that for which you lack adequate evidence.
Relation
Be relevant
Manner
Be perspicuous
1.        avoid obscurity of expression.
2.        avoid ambiguity.
3.        be brief
4.        be orderly
These maxims are meant to describe what happens in conversations and can be compared to unwritten laws.
Violation of the maxims
The use of the term principle and maxim does not mean that the CP and its maxims will be followed by everybody all the time. People do violate them and tell lies. But Grice first distinguishes telling lies from other types of violations. In his view conversational implicature can only be worked out on the basis of the CP. If somebody deliberately tells lies, the basis for accurate interpretation is lost, so lies are not implicature proper.
Examples of the violation of the maxims
(1) A professor wrote a reference letter for his past student X who is applying for a lectureship in philosophy. The letter reads:
Dear sir,
Mr. X’s command of English is excellent, and his attendance at tutorials has been regular. Yours,
XXXX.
The professor violated the maxim of quantity, implicating that X is not good at philosophy.
(2) A: Where does C live?
B: Somewhere in the south of France.
(3) Boys are boys and war is war.
(4) A: Where’s X?
B: He’s gone to the library. He said so when he left.
(5) He is made of iron.
(6) Every nice girl loves a sailor.
(7) At a genteel tea party,
A: Mrs. X is an old bag.
B: The weather has been quite delightful this summer, hasn’t it?
(8) At home,
A: Let’s get the kids something.
B: OK, but I veto I-C-E C-R-E-A-M.
Characteristics of implicature
1)        calculability （可计算性）
Implicature is calculable in that it can be worked out on the basis of previous information, including: 1) the conventional meaning of the words, 2) the CP and its maxims, 3) the context, 4) other items of the background knowledge, 5) the fact that all relevant items falling under the previous headings are available to both participants and both participants know or assume this to be the case.
2)        Cancellability （可删除性）
Also known as defeasibility. A conversational implicature relies on a number of factors as discussed in the part for calculability. If any of them changes, the implicature will also change. Example (4) above.
3)        non-detachability（非可分离性）
A conversational implicature is attached to the semantic content of what is said, not to the linguistic form. Therefore it is possible to use a synonym and keep the implicature intact. For example, John’s a genius and John’s a mental prodigy said ironically both implicate that John’s an idiot. And example (7) above. But conversational implicatures related to the Manner maxim are an exception. Example (8) above.
4)        non-conventionality（非规约性）
conversational implicature is by definition different from the conventional meaning of words. To show this more clearly, we can have a look at some examples of entailment (a logical relationship between two sentences in which the truth of the second necessarily follows from the truth of the first, while the falsity of the first necessarily follows from the falsity of the second: I saw a boy and I saw a child):
John has three cows.
John has only three cows. (implicature)
John has some cows. (entailment)
John has some animals. (entailment)
John has something. (entailment)
Somebody has three cows. (entailment)
Somebody has some cows. (entailment)
Somebody has some animals. (entailment)
Somebody has something. (entailment)
Entailments are part of the conventional meaning. If you do not know the entailment of a word you have to look it up in a dictionary. There is no way to work it out on the basis of the CP and the context. Entailment is constant in all contexts. In this sense entailment is determinate. However, implicature varies with context and therefore indeterminate.
Finally, conversational implicature and illocutionary force are both concerned with the contextual side of meaning. These two theories differ only in the mechanisms they offer for explaining the generation of contextual meaning.
Post-Gricean developments
The theory of conversational implicature has opened a new way of explaining the use of language. However, there is some inconsistency and redundancy among the CP and its maxims. Linguists of the post-Gricean period, therefore, tried to seek to boil down the maxims to a set of principles, which are truly indispensable and do not overlap at the same time.
Relevance theory
Relevance theory was proposed by Sperber and Wilson in their book Relevance: Communication and Cognition in 1986. They argue that all Gricean maxims, including the CP itself should be reduced to a single principle of relevance.
The assumption, or principle, underlying relevance theory is that in any given context, what people say is relevant. This principle of relevance can be seen as a further specification of the Gricean notion of cooperation. For relevance theory, achieving successful communication by way of the relevance of what is being said is a sufficient aim in conversation or other verbal interaction. RT is thus a minimalist theory of communication. Relevance is all we need. According to Sperber and Wilson, successful communication occurs when the speaker is recognized by the interlocutors as one who has something to say that matters or is relevant. The successful communicator is one who makes his or her intention to convey information manifest to both himself or herself and his or her partners. The mutually manifest assumption of an informative intention is at the core of Sperber and Wilson’s thinking.
This assumption, moreover, is a central trait of human communicative behavior, even to the point that one cannot properly maintain that one follows the principle of relevance: communicators do not follow the principle of relevance, and they could not violate it even if they wanted to.
According to Sperber and Wilson’s presumption of relevance, we are ready for something that makes sense (is relevant), and we will build our understanding around that assumption. The utterance we are hearing or reading is ‘accessible’ as part of our mutually recognized, common cognitive environment or context; as such the utterance is relevant. By contrast, we are not equally ready for something that would not be easily accessible, because it does not belong in such a common, cognitive environment. For example,
George has a big cat.
Here the most common assumption is that we are talking about an oversized animal of the species of genus Felis Domestica, in order to arrive at the interpretation of tiger, lion or puma, we must add something to that common environment, making it more specific (e.g. by building on the contextual information that George delivers wild animals to zoos and circuses). This latter interpretation is less relevant, because under normal circumstances, it is less accessible.
The Q- and R-principles
This is a less reductionist, bipartite model. These two principles developed by Horn, were first proposed in his “Toward a New Taxonomy for pragmatic inferences: Q-Based and R-Based implicatures” of 1984, and further elaborated in his “Pragmatic theory” of 1988.
Horn focuses on a central problem in conversational cooperation: some utterances, on a certain reading, have a clear and unambiguous meaning, while other interpretation requires a special effort on the part of the listener, e.g.
(1) I cut a finger yesterday. (whose finger is it?)
(2) I cut a nose yesterday. (whose nose is it?)
(3) I’m meeting a woman tonight. (who is the woman?)
(4) A: Wilfred is meeting a woman for dinner tonight.
   B: Does his wife know about it?
   A: OF COURSE she does. The woman he is meeting IS his wife.
Two utterances that are alike in structure can still have diametrically different meanings.
The two principles Horn introduces are: the Q-principle (Q for quantity), telling us to say as much as we can (make your contribution sufficient) and the R-principle (R for relation), which says that we should say no more than we must (make your contribution necessary). (1) invokes the R-principle to establish the fact that the finger is mine (I needn’t say more). (4) invokes the Q-principle to establish the fact that it is not his wife or regular girlfriend he’s seeing. This difference becomes a bit clearer when we think of possible replies to both questions:
(1)        A: I cut a finger yesterday.
B: That’s too bad. (Not much more can be said about the subject)
(2)        A: Wilfred is meeting a woman for dinner tonight.
B: Really? And who is she? (More will be said about the subject.)
Using Horn’s terminology, in the Q-case, I have provided as much information as I have to or can, given the circumstances; in the R-case, I let the circumstance speak, and give out only the relevant information. In this system, the Q-principle covers the Gricean maxim of quantity 1 (make your contribution as informative as is required), whereas the R-principle contains quantity 2 (do not make your contribution more informative than is required), plus the maxims of manner and relation. As to the maxim of quality, Horns leaves it alone, since, as he says, we need that in any case unless we want to see “the entire conversational … apparatus collapse.”

rossy0622

Chapter 12 Theories and Schools of Modern Linguistics
I Structural linguistics
1 Saussure
Modern linguistics began from the Swiss linguist Ferdinand de Saussure (1857-1913), often described as the father of modern linguistics.
During the years between 1907 to 1911, Saussure lectured on general linguistics in the University of Geneva. After his death in 1913, two of his students, C. Bally and A. Sechehaye, collected lecture notes from fellow students and put them together to produce the great work: Course in General Linguistics in 1916. This book became the most important source of Saussure’s ideas about linguistics and had profound influence upon succeeding generations of linguistics and marked the beginning of modern linguistics.
The important distinctions made by Saussure in linguistics:
Langue and parole
Synchronic and diachronic linguistics
Descriptive and prescriptive linguistics
Chain and choice relations
Saussure believed that language is a system of signs. This sign is the union of a form or the material sign and a idea or the conceptual sign, which Saussure called the signifier（能指） and the signified（所指）. They are both of the mental nature. The signifier is realized phonetically or graphetically. The relationship between the two is one of arbitrariness. Though we may speak of the signifier and the signified as if they were separate entities, they exist only as components of the sign. The sign is the central fact of language.
2 The Prague school
The Prague School (Circle of Linguistics of Prague) practiced a special style of synchronic linguistics and its most important contribution to linguistics is that it sees language in terms of function.
Phonology and phonological oppositions
The Prague School is best known and remembered for its contribution to phonology. Trubetzkoy declares, “the phoneme can be defined satisfactorily neither on the basis of its psychological nature nor on the basis of its relation to the phonetic variants, but purely and solely on the basis of its function in the system of language.
Phonemes are capable of distinguishing between different words because they themselves are different from one another. This difference between phonemes is called phonological opposition in the Prague School.
In their discussion of phonological opposition, they notice that some oppositions are effective in all contexts and some are effective only in certain contexts. They call the former type constant opposition and the latter neutralizable opposition. For example, in English, /p/ and /b/ are two distinctive sounds; substituting one for the other generally changes the meaning of a word. But in the context of /s-/, this opposition is lost, i.e. neutralized. No two English words are distinguished by the presence of /sp/ in one and the presence of /sb/ in the other. There is only one sound occurring after /s/, whether it is grouped with the phoneme /p/ or /b/.
In Prague School phonology, a phoneme is defined by the opposition it enters. Since the opposition is different when it is neutralized from when it is not, the phoneme in the neutralized position cannot be the same as the phoneme in the non-neutralized position. That is to say, the English phoneme should not be the same as either /p/ or /b/. Prague School linguists use a special symbol for it - the capitalized /P/, and term it ARCHIPHONEME.
Functional sentence perspective
According to the Prague school linguists, apart from the analysis of a sentence in terms of subject and predicate from the formal point of view, there may also be a functional analysis in terms of Theme (the starting point of an utterance) and Rheme (the nucleus or the core of the utterance). In English, theme and rheme are often expressed by the grammatical subject and predicate.
I went to the zoo yesterday.
However, there are utterances in which they do not correspond, e.g.
Yesterday I went to the zoo.
At the airport I met John yesterday.
The analysis of a sentence in terms of theme and rheme is now known as the functional sentence perspective because this patterning is determined by the functional approach of the speaker.
The London school
The London School refers to the kind of linguistic scholarship in England, a country that has both an unusually long history in linguistics and peculiar features in modern linguistics. The man who turned linguistics proper into a recognized distinct academic subject in Britain was J. R. Firth (1890 - 1960), the first Professor of general linguistics in Great Britain (1944). Firth was influenced by the anthropologist B. Malinowski (1884 - 1942). In turn, he influenced his student, the well-known linguist M. A. K. Halliday. The three men all stressed the importance of context of situation and the system aspect of language. Thus, London School is also known as systemic linguistics and functional linguistics.
Malinowski’s theories
Malinowski was Professor of Anthropology at the London School of Economics from 1927 onwards. The most important aspect of his theorizing, as distinct from his purely ethnographic work, concerned the functioning of language. For Malinowski, to think of language as a means of transfusing ideas from the head of the speaker to that of the listener was a misleading myth. He said that language is to be regarded as a mode of action, rather than as a counterpart of thought.” According to him, the meaning of an utterance does not come from the ideas of the words comprising it but from its relation to the situational context in which the utterance occurs.
Malinowski believed that utterances and situation are bound up inextricably with each other and the context of situation is indispensable for the understanding of the words. The meaning of spoken utterances could always be determined by the context of situation. He distinguished three types of context of situations:
(1)        situations in which speech interrelates with bodily activity: the meaning of a word is not given by the physical properties of its referent, but by its functions. When a savage learns the meaning of a word, the process is not accompanied by explanation but by learning to handle it. Likewise, a verb, a word for an action, receives its meaning through an active participation in the action.
( 2 ) narrative situations: Malinowski further distinguished “the situation of the moment of narration” and “the situation referred to by the narrative”. The first case is “made up of the respective social, intellectual and emotional attitudes of those present, and the second case derives its meaning from the context referred to (as in a fairy tale).
( 3 ) situations in which speech is used to fill a speech vacuum – phatic communion: cases of  language used in free, aimless, social intercourse, such use of language is not the least related to human activities, and its meaning cannot possibly come from sociability and … the fact of the personal communion of these people. For example, the function of a polite utterance has nothing to do with the meaning of the words in it. Malinowski called such utterances “phatic communion”.
Firth’s theories
Influenced by Malinowski, Firth regarded language as a social process, as a means of social life, rather than simply as a set of agreed-upon semiotics and signals.
He insisted that the object of linguistic study is language in actual use. And the goal of linguistic inquiry is to analyze meaningful elements of language in order to establish corresponding relations between linguistic and non-linguistic elements. The methods of linguistic study is to decide on the composite elements of language, explain their relations on various levels and ultimately explicate the internal relations between these elements and human activities in the environment of language.
Firth held that meaning is use, thus defining meaning as the relationship between an element at any level and its context on that level.
According to his theorizing, the meaning of any sentence consists of the following five parts:
(1) the relationship of each phoneme to its phonetic context;
(2) the relationship of each lexical item to the others in the sentence ;
(3 ) the morphological relations of each word;
(4) the sentence type of which the given sentence is an example;
(5 ) the relationship of the sentence to its context of situation.
Firth’s own study focused on the context of situation as Malinowski did. He defined the context of situation as including the entire cultural setting of speech and the personal history of the participants rather than as simply the context of human activity going on at the moment.
Firth tried to set up a model for illustrating the close relationship between language use and the context of situation which contains the following components:
A.        the relevant features of the participants: persons, personalities
(i)        The verbal action of the participants
(ii)        The non-verbal action of the participants
B.        The relevant objects
C.        The effects of the verbal action
Firth made more specific and more detailed contextual analyses. He put forward the idea that in analyzing a typical context of situation, one has to carry out the analysis on the following four levels :
( 1 ) The internal relations of the text
( a ) the syntagmatic relations between the elements in the structure;
(b) the paradigmatic relations between units in the system and find their values.
2) The internal relations of the context of situation
(a) the relations between text and non-linguistic elements, and their general effects;
(b) the analytical relations between “bits” and “pieces” of the text (words, parts of words, phrases) and the special elements within the situation (items, objects, persons, personalities, events).
Halliday and the systemic-functional grammar
M. A. K. Halliday ( 1925-) has developed the idea stemming from Firth’s theories in the London School. His Systemic- Functional (SF) Grammar is a sociologically oriented functional linguistic approach.
Systemic-Functional Grammar has two components: SYSTEMIC GRAMMAR and FUNCTIONAL GRAMMAR.
Systemic grammar
The system is a list of choices that are available in the grammar of a language. The number system in English, for example, contains two choices: singular and plural. The person system offers three choices: first person, second person, and third person. There are also systems of gender, tense and mood, etc.
A SYSTEM is a list of things between which it is possible to choose. So they are meanings, which the grammar can distinguish. The items in a system are called options. And the items of a particular system have something in common, belonging to the same area of meaning. For instance, singular and plural are distinct, but they both have to do with number. All systems have three essential characteristics.
Firstly, the terms in a system are mutually exclusive.
Secondly, a system is finite. It is possible to fix a limit for a system and to say that it consists of a certain countable number of items, no more, and no less .
Thirdly, the meaning of each item in a system depends on the meaning of the other items in the system. If the meaning of one term is changed, the meaning of other items will also change.

                                     yes / no
                     interrogative®    wh
          indicative®                 assertion
                     declarative ®    exclamation
Clause®                                    
          imperative®                                                                                   
Functional grammar
functional aspect, is also termed as sociosemantics. According to Halliday, the context of situation or the social context contains three components:
Field of discourse: the subject matter being discussed.
Tenor of discourse: the social relations between the participants of conversations
Mode of discourse: the channel of communication
These three situational components are related to the three important functions of language identified by Halliday: ideational, interpersonal and textual: language serves for the expression of content (ideational); language serves to establish and maintain social relations (interpersonal); language enables people to construct texts (textual). These three functions are related to three grammatical systems: transitivity, mood and theme.
The relationship between the three situational components, the three language functions and the three grammatical systems can be represented as follows:
                  Phonology / graphology
Transitivity: the 6 processes        Mood-residue structure        Theme-rheme structure
Ideational meaning        Interpersonal meaning        Textual meaning
Field of discourse        Tenor of discourse        Mode of discourse
Transitivity: the representation of outer experiences and inner experiences.
Material processes: processes of doing
Actor: the one who does the deed
Goal: the one the deed is extended to or directed at
  The lion        caught        the tourist
  Actor          Process           Goal
Mental Processes: Processes of sensing
Senser: the one who feels
Phenomenon: what is being felt
  He          saw          the morning star
  She          liked          the gift
  I          know          they don’t care
  I          don’t understand        it
Senser         Process                    phenomenon
Relational Processes: Processes of being
Sarah   is    wise
Carrier Process Attribute
Tom      is      the leader
Identified  Process  Identifier
Behavioural Processes
These are processes of (typically human) physiological and psychological behabiour.
Behaver: the participant, typically a conscious being, who is behaving,
No-one’s listening.
He’s always grumbling.
Verbal Processes: processes of saying
Sayer: the one (not necessarily a human being: my watch says it’s seven) who says, e.g. John in John said he’s hungry.
Receiver: the one to whom the saying is directed, e.g. me, your parents in Tell me the whole truth. and Did you repeat that to your parents?
Verbiage: the function that corresponds to what is said
The manager will outline his plan of campaign.
Existential Processes
These represent that something exists or happens, e.g.
There was a little pig.
There isn’t enough time.
Mood structure: the representation of exchange
The nature of dialogue
        Commodity          exchanged

Role in
exchange
       
(a) goods-&-service

       
(b) information




(i) giving
        ‘offer’
Would you like this teapot?
        ‘statement’
He’s giving her the teapot.


(ii) demanding        ‘command’
Give me that teapot        ‘question’
What is he giving her
Mood consists two parts: the subject (a nominal group) and the finite operator (part of the verbal group expressing tense or modality).
The duke has given away that teapot, hasn’t he?
Residue: the remainder of the clause: predicator, complement, adjunct:
Sister Susie        ’s        sewing        shirts        for soldiers.
Subject        Finite        predicator        Complement        Adjunct
Mood        Residue
1)        the presence of the mood element, consisting of Subject plus Finite, realizes the feature ‘indicative’;
2)        within the indicative what is significant is the order of Subject and Finite:
a)        the order Subject before Finite realizes ‘declarative’
b)        the order Finite before Subject realizes ‘yes-no interrogative’
c)        in a ‘wh-interrogative’ the order is
i)        Subject before Finite if the wh-element is Subject
ii)        Finite before Subject otherwise
Theme and rheme
Theme: the element which serves as the point of departure; it is that with which the clause is concerned. As a general guide, the Theme can be identified as the element which comes in first position in the clause.
Rheme: the part in which the theme is developed, the remainder of the message, is called the Rheme.
A message consists of a Theme combined with a Rheme.
e.g.
  once upon a time
  very carefully
  for want of a nail
  with sobs and tears              there were three bears
      she put him back on his feet again
      the shoe was lost
       he sorted out those of the largest size
  Theme              Rheme
Given vs new information
Given information: information that is presented by the speaker as recoverable
New information: information that is presented by the speaker as not recoverable
John sat in the front seat.
So what did you say?
Are you going to paint the kitchen blue?
In this job, Anne, we are working with silver now. Silver needs to have love.
I’ll tell you about silver. It needs to have love.
I’ll tell you what silver need to have. It needs to have love.
According to Halliday, a clause is the simultaneous realization of ideational, interpersonal, and textual meanings. For example,
Ideational        This house           was built                 by John Smith
Goal                Process: material          Actor
Interpersonal        Subject     finite operator Predicator            Adjunct
Mood                  Residue
textual        Theme              Rheme
Given                                      New
American Structuralism
American Structuralism is a branch of synchronic linguistics that emerged independently in the United States at the beginning of the twentieth century. It developed in a very different style from that of Europe, under the leadership of anthropologists and linguists as F. Boas, E. Sapir, L. Bloomfield, Z. Harris, C. Hocket, whose tradition actually influenced the whole of the 20th century American linguistics.
Boas and Sapir
Unlike the Europeans who guessed the universals of language, Boas held that there were no ideal types or forms of languages, for human languages were endlessly diverse. Although the structure of a language in some primitive tribes might sound very arbitrary and irrational, there was no basis of such a judgment, because European languages would appear just as irrational to a member of those tribes. Boas as strongly opposed to the view that language is the soul of a race, and he proved that the structure and form of a language has nothing to do with the evolution of a race and the development of a culture.
Boas discussed the framework of descriptive linguistics. He held that such descriptions consist of three parts: the sound of languages, the semantic categories of linguistic expression, and the process of grammatical combination in semantic expression. Boas noticed that every language has its own system of sounds and its own grammatical system. He held that the important task for linguists is to discover, for each language under study, its own particular grammatical structures and to develop descriptive categories appropriate to them. His methodology in processing linguistic data of American Indian languages is analytical, without comparing them with such languages as English or Latin. Starting from an anthropological view, Boas regarded linguistics as a branch of anthropology and failed to establish linguistics as an independent branch of science.
Like Boas, Sapir (1884 - 1939) was an eminent anthropological linguist. Influenced by Boas, Sapir undertook the description of American Indian languages, using a native informant in his own cultural surroundings. This is a novel experience for Sapir and radical departure from the traditional practice of trying to impose the grammatical categories from the Indo-European languages upon all other languages. His idea on language and thought was later developed by his student. B. L. Whorf (1987 – 1941) and is known as the Sapir - Whorf Hypothesis.
Bloomfjeld
The principal representative of American descriptive linguistics is L. Bloomfield (1887 - 1949). He is such a landmark figure in the history of American linguistics that the period between 1933 and 1950 is known as Bloomfieldian Era, in which American descriptive linguistic formally came into being and reached its prime development.
For Bloomfield, linguistics is a branch of psychology, and specifically of the positivistic branch of psychology known as BEHAV1OURISM. Behaviourism is a principle of scientific method, based on the belief that human beings cannot know anything they have not experienced. Behaviourism in linguistics holds that children learn language through a chain of stimulus-response reinforcement, and the adult’s use of language is also a process of STIMULUS-RESPONSE.  When the behaviourist methodology entered linguistics via Bloomfield’s works, the popular practice in 1iguistic studies was to accept what a native speaker says in his language and to discard what he says about it. This is because of the belief that a linguistic description was reliable when based on observation of unstudied utterances by speakers. It was unreliable if the analyst had resorted to asking speakers questions such as “Can you say .. . in your language?”
Bloomfield also touched upon the application of linguistics to language teaching and criticised traditional grammar.
Post-Bloomfieldian linguistics
Influenced by Bloomfield’s Language, American linguists such as Z. Harris (1909 - ), C. Hockett (1916 - 2000) further developed structuralism. characterlsed by a strict empiricism.
Post-Bloomfieldian linguistics focused on direct observation: a grammar is discovered through the performing of certain operations on a corpus of data, i.e. through discovery procedures. The corpus of data consists of speech, so the operation has to start from a phonological analysis of the stream of sounds as phonemes. Since phonemes form a variety of types of structures, they can be grouped into minimal recurrent sequences, or morphs, which are the members of the same morphemes . Based on the discovery of morphemes of the language, the task of the linguist is to discover how the morpheme may be combined in order to produce a grammar. The post – Bloomfieldian linguists also took an interest in the discourse level in order to develop discovery procedures for structure above the sentence level.
Hockett was both a linguist and anthropologist, remaining firmly within the structuralist paradigm and hailed as a star of post-Bloomfieldian linguistics. He contributed a lot to the development of phonemic, morphemic and grammatical analysis and general linguistics. And he is well remembered for his discussion of the design features of language.
II. Generative-transformational linguistics (Noam Chomsky)
In the late 1950s N. Chomsky tried to open up a new route when he found that the classification of structural elements of language according to distribution and arrangement had its limitations. From this practice Chomsky gradually established the well-known transformational-Generative grammar. From its birth to the present day, TG grammar has undergone five stages of development. The classical theory aims to make linguistics a science. The standard theory deals with how semantics should be studied in a linguistic theory. The extended standard theory focuses discussion on language universals and universal grammar. The revised extended theory focuses discussion on government and binding. The latest is the minimalist program, a further revision of the previous theory.
The innateness hypothesis
Chomsky believes that language is somewhat innate, and that children are born with what he calls a language acquisition device (LAD), which is a unique kind of knowledge that fits them for language learning. He argues that children are born with knowledge of the basic grammatical relations and categories, and this knowledge is universal. And the study of language can throw some light on the nature of the human mind. This approach to language is a reaction against behaviorism in psychology and empiricism in philosophy, making linguistics a branch of psychology.
Chomsky’s innateness hypothesis is based on his observations that some important facts can never be otherwise explained adequately:
1)        Children learn their native language very fast and with little effort.
2)        Children learn their mother tongue in very different environments, but the result is roughly the same.
3)        Children learn the total grammar of the language during a limited period of time, from limited exposure to speech.
All these suggest that although babies are not born knowing a language, they are born with a predisposition to develop a language in much the same way as they are born with a predisposition to learn to walk.
Chomsky argues that LAD probably consists of three elements: a hypothesis-maker, linguistic universal, and an evaluation procedure. Children are born into many different speech communities and, with equal ease, can pick up the language of the community they happen to be born into. They look for regularities in the speech around them, then make guesses on the basis of more linguistic data and make more hypotheses. To do all this, they need the hypothesis-maker in LAD. However, there are times when two or more hypotheses cover the same set of linguistic facts, but one of them is simpler and better. In other words, there might be several grammars that can account for the data children are exposed to. How is it that children always choose the better hypothesis or the better version of grammar? Chomsky suggests that to cope with the problem children must be equipped with an evaluation procedure, which will allow them to choose between a number of possible grammars.
The classical model (1957: Syntactic Structures)
deep and surface structures
deep structures: the abstract representation of the syntactic properties of a construction.
surface structures: the final stage in the syntactic derivation of a construction, which closely corresponds to the structural organization of a construction people actually produce and receive.
For example, in its surface structure Children are hard to please, children is the subject and to please is part of the complement. But in its deep structure, is hard would have a subordinate sentence as its subject and in the subordinate sentence children is the object of please: thus, in outline, s[please children] is hard. In contrast, though in its surface structure Children are eager to please, children is still the subject and to please part of the complement, however, in its deep structure, to please would have a subject children and an object others: thus, children are eager s[children please others]
A graphic presentation of the classical model:
Phrase structure component
Deep structure
Transformational component
Surface structure
Morpho-phonemic component
phrase structure rules
The phrase structure component has phrase structure rules. e.g.
S→NP +VP
VP→V + NP
NP→Det+ N
Det→the, a, etc.
N→man, ball, etc.
V→hit, took, etc.
                            S

                 NP                VP

          Det         N      V             NP


          The        man  hit        Det          N


                                       a         ball
                S

           NP         VP

           N      V        NP

          John   likes N           S

                   peaches NP         VP

                          Pron   V         NP

                          that  have   Adj      N

                                      smooth  skin
transformational rules
The transformational component has transformational rules which change the deep structures into surface structures. According to Chomsky, active and passive, declarative and interrogative, positive and negative sentences have the same deep structures. The difference between them simply comes from the operation of the relevant transformations, e.g. the deep structure of John will finish the paper tomorrow is something like:
NP1 + Aux + V + NP2
The passive transformation:
NP2 + Aux + be + en + V + by + NP1
The paper will be en finish by John (tomorrow)
John gave a book to Mary.
NP1 + V +NP2 + to + NP3
The dative movement transformation:
NP1 + V + NP3 + NP2
John + gave + Mary + a book
The morpho-phonemic component is responsible for the correct spelling and pronunciation of the words in the surface structure, e.g. the affix hopping transformation will change the order of en + finish and a morpho-phonemic rule will get en to be spelt as ed and pronounced as /t/.
The standard theory (1965: Aspects of the Theory of Syntax)
*Colorless green ideas sleep furiously.
The graphic presentation
Base Component
Categories  Lexicon
Deep Structure                             Semantic Component
Transformational Component
Surface Structure
Phonological Component
The base and transformational components together form the syntactic component. So now language is seen as consisting of three major parts: syntax, semantics and phonology.
Categories contain rewriting rules more or less the same as the phrase structure rules in the classical model. But at the end of a derivation there are no longer rules inserting words directly, but rules with feature specifications for the words to be inserted.
Features of nouns: + Common, + Count, + Abstract, + Animate, + Human
Features of verbs: + --- #, + --- NP, + --- Adjective, + --- that⌒S’
e.g. sincerity [ + N, (+ Common,) (- count,) + Abstract]
   boy [ + N, + Common, (+ Count,) (+ Animate,) + Human]
   elapse [ + V, + --- #]
   grow [ + V, + --- NP, + --- #, + --- Adjective]
A more complete specification will have to include the kind of subject or object a verb takes as well, e.g. Frighten→[ + V, + --- NP, + [ + Abstract] Aux --- Det [ + Animate], …], Sincerity may frighten the boy.
Sub-categorization of this type is intended to ensure the generation of only acceptable sentences.
The extended standard theory (Studies on Semantics in Generative Grammar)
I have been taught physics by Newton.
Newton has taught me physics.
Beavers build dams.
Dams are built by beavers.
Surface structures also have some bearing on semantic interpretation. So the standard theory was extended as:
Base Component
Categories  Lexicon

Deep      Structure                            Semantic Component
Transformational Component
Surface     Structure



Phonological Component
The trace theory
It was suggested that if we introduce the notion trace, all the necessary information for semantic interpretation may come from the surface structure. That is after the movement of an element there will be a trace left in the original position, e.g.
Beavers build dams.
Dams are built by beavers.
Dams are built t by beavers.
The deep structure information that dams was originally the object of build is also captured by the surface structure. Then the semantic component could rely entirely on the output of surface structure.


Base Component
Categories  Lexicon

Deep      Structure                     


Transformational Component
Surface  Structure


Phonological Component
Semantic Component