Notes on Language, Writing, Prescriptive Grammar, and Modality (1.3–1.5)
1.3.1 What Language Inherently Is and Is Not
Language knowledge (competence) = vast mental knowledge about using a language to communicate ideas.
Non-essential aspects of language that can cloud what it means to know a language: writing and prescriptive grammar.
Goal: show that writing and prescriptive rules are related but not fundamental to “what you know” when you know a language.
Writing and prescriptive grammar are not primary focus of linguistics (though Chapter 15 discusses writing systems).
Writing and prescriptive rules often have prominent roles in other classes about language(s).
1.3.2 Writing Is Secondary to Speech (And Not Necessary for Knowledge of a Language)
Speaking/signing vs writing are two forms of communication with different functions; neither is superior. Writing is not a more perfect form of communication.
Language, as in File 1.2, consists of mental knowledge: a lexicon and a mental grammar.
To reveal linguistic knowledge, speakers must perform language; speech is the primary manifestation of language.
Modern linguistics treats speech (spoken or signed) as the primary object of study; writing is the representation of language in a different physical medium.
Speech is considered primary because it is more immediate; writing adds a step: idea → mental grammar → written form.
Modern technology blurs the line (texting, email, recordings can preserve speech; recordings can preserve speech; text can be transmitted quickly).
For data, linguists prefer spoken language as the best source of data and object of description (except for languages with no native speakers, e.g., Latin).
When illustrating examples, conventional written transcriptions of audio forms are used, with phonetic conventions given in Chapter 2; audio icons indicate recordings (see File 1.1.5).
Reasons why speech is more basic than writing:
a. Writing must be taught; spoken language is acquired naturally. Children learn to speak naturally; writing requires explicit teaching.
b. Writing does not exist everywhere spoken language exists
c. Neurolinguistic evidence shows that writing processing/production overlays brain areas involved in speech.
d. Writing can be edited before sharing, while speech is often spontaneous; this shows immediacy of speech.
e. Writing arose later historically than spoken language (writing in Sumer ~6,000 years ago; spoken language likely hundreds of thousands of years).
Misconception: writing is more perfect than speech. Several reasons contribute:
a. Writing can be edited and thus appears more polished; writing often results from deliberation and revision; speech is spontaneous.
b. Writing is associated with education and the standard variety; this can give the impression that writing is the norm, though standard writing is not inherent to language.
C. Writing is more physically stable (written forms persist) and spelling appears less variable than pronunciation; however, this stability does not indicate that writing is more fundamental.
Note: linguists focus on spoken language; writing relates to language in fascinating ways (Chapter 15).
1.3.3 Language Is Not Prescriptive Grammar
Three kinds of grammar:
(a) mental grammar: the linguist’s actual understanding in individuals or groups.
(b) descriptive grammar: the linguist’s description of how a language is spoken.
(c) prescriptive grammar: socially embedded notions of what is “correct” or “proper.”
Descriptive grammar describes actual usage without judgment; prescriptive grammar judges usage as “correct” or “incorrect.”
Prescriptive rules illustrate what people think should happen, not what actually happens. Examples of prescriptive rules (1)–(1c):
a. Do not end a sentence with a preposition. NO: Where do you come from? YES: From where do you come?
b. Do not split infinitives. NO: to boldly go where no one has gone before; YES: to go boldly where no one has gone before
c. Do not use double negatives. NO: I don’t have nothing. YES: I don’t have anything. I have nothing.
Prescriptive rules make value judgments about correctness; but mental grammar rules are what actually exist and can differ across speakers.
Descriptive statements illustrate language competence without normative judgments:
a. Some English speakers end sentences with prepositions.
b. Some English speakers split infinitives.
c. Some English speakers use double negatives for negation.
Descriptive grammars may vary by speaker groups (e.g., Ohio State University undergraduates may produce constructions like The room needs painted, which are grammatical for some speakers but not for others — adaptions occur over time as mental grammars learn from exposure).
Prescriptive rules may persist because they are linked to social status and norms; nonstandard dialects can be stigmatized, affecting social mobility and identity.
Language varieties: prescriptive rules are used to mark social identity and mobility, but they do not reflect intrinsic linguistic validity. Descriptive grammars remain the tool for discovering mental grammars.
Historical origins of prescriptive rules (seventeenth–eighteenth centuries) include idealization of classical Latin as a pure standard; several prescriptive rules mirrored Latin rules rather than actual English use:
(1a) Endings with prepositions were discouraged because Latin did not end with prepositions.
(1b) Infinitives should not be split because Latin infinitives were single words.
(1c) Double negatives were viewed as nonlogical; this was extended from logic to language.
Old English permitted double negatives; by Shakespeare’s time, double negatives were uncommon in educated speech but common in dialects. Bishop Lowth (1762) argued against doubles by logic, but usage-based descriptivism shows that many languages permit doubles; language change continues regardless of prescriptive rules.
Prescriptive rules persist because they correlate with social status and standard language ideals, but they are not linguistically superior. Linguists rely on descriptive grammars to understand mental grammars.
Change over time: ideas about grammaticality can change; Jane Austen’s 1818 example (The trunk is being carried down) shows that a construction once considered awkward may become acceptable.
1.4.1 How to Identify Language When We Come across It
Defining language is difficult; one approach is to identify descriptive features that characterize language.
Hockett’s design features of language (the design features) describe descriptive properties of language; there are nine such features. While many communication systems share some features, only those displaying all nine features can be called a language. The features are ordered from most universal to most specific; first three are shared by all communication systems; final two are unique to human language.
The standard version of Hockett’s design features provides a framework for identifying languages and distinguishing natural languages from other forms of communication.
1.4.2 Mode of Communication
Mode of communication = the means by which messages are transmitted and received.
Most human languages are auditory-vocal (spoken) but many are visual-gestural (signed).
The term language modality refers to the production and perception channel (audio vs visual).
Sign languages are acquired as first languages in childhood and can be learned later; they are not mere codes or pantomime.
1.4.3 Semanticity
Semanticity means that signals have meaning or function within a system; words/signs convey intended meaning.
Examples: even unknown words/sentences carry meaning; e.g., understanding that There was a large amount of frass in the tubes with the fruit flies has meaning even if frass is unknown.
1.4.4 Pragmatic Function
Language serves a purpose beyond meaning: it can help us stay alive (ask for food, request help), influence others, and learn about the world.
Sets phrases (e.g., greetings like "Nice weather today" or "Hey, what’s up?") help initiate conversation and maintain social structure.
Gossip also serves social bonding and information exchange.
1.4.5 Interchangeability
The ability of individuals to both transmit and receive messages: everyone can produce and understand language.
1.4.6 Cultural Transmission
Language aspects are learned through social interaction, not solely genetically predetermined.
Children learn the language(s) or dialect(s) used by their community; lack of exposure to language in childhood prevents language acquisition.
1.4.7 Arbitrariness
a. Linguistic Sign: Form + Meaning = Linguistic Sign. Example: pit [pɪt] = inner core of a peach (pit) in English. The form–meaning connection is typically arbitrary; the same form can have different meanings in different languages, and different forms can express the same meaning.
b. Evidence for Arbitrariness: cross-linguistic differences; the same meaning often has different sounds across languages; there are no universally recognized forms for given meanings.
c. Onomatopoeia: some forms imitate natural sounds, but even onomatopoetic words vary across languages (e.g., rooster sound in English [kɑkədudldu] vs Chinese [kukuku]); cross-linguistic tables show that onomatopoeia is not universally identical.
d. Sound Symbolism: some sounds evoke meanings (e.g., high-front vowels like [i] in words meaning smallness: teeny, petite, wee, mikros, -ito); these show partial nonarbitrariness due to phonetic associations with meanings.
e. Nonarbitrary Aspects of Language: while arbitrariness dominates, there are modest nonarbitrary aspects (e.g., poets manipulate onomatopoeia for phonetic effect; sound symbolism can influence poetic effect).
Notes: nonarbitrariness and iconicity are limited in scope; they are present but secondary in most language aspects.
1.4.8 Discreteness
He is fast is built from discrete units: words and individual sounds; sounds like [h], [i], [ɪ], [z], [f], [æ], [s], [t] are discrete units.
A language has a limited inventory of sounds (roughly 10–100); e.g., English has about ~50 sounds.
These discrete units can be recombined to form many meaningful units (words), and words can be rearranged to form larger units (phrases, sentences).
From a small set of meaningless units, a large number of meanings can be created (potentially infinite).
1.4.9 Displacement
The ability to talk about things not present in space/time during the act of communication (e.g., color red not present; people in different states; past/future events; fictional beings).
1.4.10 Productivity
Productivity = language’s capacity to generate novel messages from discrete units; there is no fixed set of possible combinations.
Humans can produce and understand an infinite number of sentences; novel sentences convey new propositions (e.g., "Funky potato farmers dissolve glass").
Understanding of new sentences relies on knowledge of the rules for combining words and their meanings.
All levels of linguistic structure have productive rules; these rules generate new forms for communication.
1.4.11 What the Design Features Tell Us, and What They Don’t Tell Us
All languages exhibit all nine design features; any system lacking them cannot be a language; only human communication systems display all nine features (File 14.1 discusses animal communication in relation to these features).
A system that displays all nine features is not automatically a natural language. Examples include formal languages (e.g., C++, mathematical logic) that cannot be acquired as a native language by children.
Constructed languages (e.g., Esperanto, Elvish, Klingon) may imitate some properties of natural languages but are not necessarily natural languages. Distinctions are drawn between:
natural languages: evolve naturally in a speech community; learned by children; have evolving lexicons/grammars.
constructed languages: invented and may or may not become natural in a community; may aim for international communication or fictional purposes.
formal languages: not learnable by children as a native language; not acquired via community usage; different from natural languages.
For purposes of the book, the focus is on natural languages; other types exist but are distinguished.
1.5.1 Auditory-Vocal and Visual-Gestural Languages
Language is a cognitive system with rules in the brain; however, transmission depends on modality: how it is produced and perceived.
Most familiar languages are auditory-vocal (spoken); others are visual-gestural (signed).
Signed languages are real languages with vocabulary and grammar; they can be learned in childhood or later; they share core linguistic properties with spoken languages.
1.5.2 Some Common Misconceptions about Visual-Gestural Languages
a. Signed Language vs. Manual Codes: myths that signed languages derive from spoken languages or are mere codes. Codes (e.g., Morse) borrow structure from spoken languages; signed languages evolve independently with distinct structure; they have native signers and are not just codes.
Signed languages do have native signers, whereas codes do not typically have native speakers.
See examples: SEE II (manual code for English) vs ASL; ASL is a natural language with its own grammar; SEE II mirrors English morphology and is slower (rates measured as seconds per proposition: English/ASL ≈ 1.5 s; SEE II ≈ 2.8 s).
b. Signed Language vs Pantomime: myths that signs are just pictures or pantomime; signs do have phonological/morphological/syntactic structure; the majority of signs are not purely iconic and the form-meaning relationship is largely arbitrary, as in spoken languages.
Signs can be iconic to some extent (e.g., KNOW sign with forehead contact or cheek touch), but over time signs become more arbitrary; signs can convey abstract concepts and not just concrete objects.
c. Universality of Signed Languages: myths that there is a single universal sign language; there are over 150 documented signed languages, not mutually intelligible, with differences similar to spoken languages.
1.5.3 Who Uses Signed Languages?
Signed languages are used worldwide wherever there are deaf communities; sign languages can be learned by both deaf and hearing individuals.
Martha’s Vineyard (18th–19th centuries) and the Al-Sayyid Bedouin tribe (Israel) had substantial sign usage; large deaf populations can lead to widespread signing across the community.
Deaf community (capital D) refers to Deaf culture, identity, and language use (ASL in the US); not all deaf individuals participate in Deaf culture or use sign language; some read lips or use other communication modes.
The choice of communicative mode depends on social and practical factors.
1.5.4 Representing Signs in a Two-Dimensional Format
Signs cannot be written with the Roman alphabet due to different modalities; conventions include using capitalized English words to label signs (e.g., DOG for the sign for dog).
In addition to meaning labels, signs can be shown by:
photographs of signers
drawings of people producing signs
drawings showing only the hands (without the signer)
These representations are approximations; signs cannot be captured perfectly in static two-dimensional images.
For further resources, URLs are provided to online ASL dictionaries and sign resources (e.g., Lifeprint, Handspeak, ASLPro, SignSignSAVVY).
1.5.5 The Importance of Studying Different Modalities
Some linguistic principles are expressed differently across modalities (spoken vs signed), but many similarities exist; modality differences help identify which aspects of language are universal and which are modality-specific.
Example: pause duration in spoken languages has a minimum length related to breathing and cognitive planning; pause lengths in signed languages do not show a minimum length, suggesting the min-pause length is specific to auditory-vocal modalities.
The majority of examples in this book focus on spoken languages (English) for familiarity, but cross-modal study is essential for a broader understanding of language.