Chapter 1: Human Language and Language Science – Study Notes
Chapter 1 Notes: Human Language and Language Science
- Overall aim of the chapter
- Introduce what language is and how language scientists (linguists) study it
- Move away from simple right/wrong rule-based thinking to empirical, descriptive observation of language use
- Acknowledge how language studies have intersected with colonial power and harm, and consider ethical implications
- Learning goals:
- Differentiate prescriptive vs descriptive approaches
- Identify components of mental grammar
- Explain properties common to all human languages
- Describe techniques for doing language science
- Discuss ethics of doing language science
1.1 What Even Is Language?
- Language is used by all of us; linguistics is the scientific study of human language
- Language has multiple meanings and uses beyond linguistics (e.g., computer languages, body language, love languages)
- For study, focus on a language in use (here, a variety of English)
- Modality of language
- Vocal languages: speech sounds produced by larynx, tongue, teeth, lips; heard by ears
- Signed languages: manual signs, visual/touch reception
- In this book: reserve terms speaking/speech for vocal languages; refer to language users for all modalities; occasionally use "languaging" as a verb
- The shared mental system: the mental grammar
- Goal of linguistics: figure out what this shared system is like
- The mind has a grammar that enables understanding and being understood
- How do we understand language scientifically?
- Consider what counts as talking (modality) and how to segment continuous streams of sound/gesture into meaningful units
- Example of segmentation and form-meaning connections
- Mental lexicon: links sound/gesture forms to meanings; like a dictionary in the mind
- For most words, form-meaning pairings are arbitrary (e.g., English ‘pumpkin’ vs Nishnaabemwin ‘kosmaan’)
- Some words exhibit iconic relationships between form and meaning (discussed later)
- Expressing/demanding language:
- To say “cookie,” you must know how to articulate the word or sign it; this involves articulatory phonetics
- Implicit vs explicit knowledge in language
- When you know a language well, much of grammar is implicit (unconscious)
- Lexicon and morphology (word structure) are often more conscious
- Articulatory phonetics and other phonology knowledge are typically implicit
- Parts of the grammar (conceptual overview)
- Phonology: how physical units of language combine and change in context
- Morphology: the internal structure of words (e.g., plural formation)
- Syntax: how words combine to form phrases and sentences; supports semantic interpretation
- Semantics: meaning of words and sentences
- Pragmatics: meaning in context
- Reading/writing as part of grammar
- Literacy is not required for grammatical competence; literacy is a skill that can be woven into mental grammar when learned
- Writing systems are not universal across languages (e.g., Mongolian Cyrillic vs traditional Mongolian script; some languages (e.g., ASL, LSQ) do not have a standard written form)
- Some languages have no written forms; sign languages can be fully competent without a writing system
- Literacy’s integration with grammar when present; literacy is secondary to core grammar
1.2 What Grammars Are and Aren’t
- Distinction between grammar in everyday use vs the mental grammar studied in linguistics
- Mental grammar: the system in the mind that allows understanding/production of language
- No grammar is better or worse than another; all languages/dialects are equally valid from a scientific perspective
- Descriptive vs prescriptive approaches
- Descriptive: describe how language is actually used
- Prescriptive: prescribe how people should use language
- Linguistics aims to describe, not judge or rank languages
- Creativity/productivity of grammar
- All languages can generate an infinite number of sentences from a finite vocabulary and finite set of rules
- Language change is constant over time (lexicon, phonetics/phonology, syntax, morphology, semantics)
- This productivity explains why languages evolve and why older generations often feel language is changing
- Language standards and power dynamics
- Standard languages are often associated with prestige and power due to colonial histories and social hierarchies
- There is no objective standard; standards reflect social/political power more than linguistic superiority
- Examples and implications discussed: French Académie Française; Queen’s English; Canadian/UK/American varieties; Black English (AAL/AAVE) as markers of identity and social positioning
- Consequences of standardization
- Education and policy can privilege certain varieties, disadvantaging speakers of other dialects or minority languages
- Examples include Haiti’s public education in French despite wide use of Haitian Creole; Nigerian English accents; judgements about ASL varieties in legal contexts
- Standard vs dialect vs variety terminology (to be elaborated in Chapter 2)
- Whole-language creativity and change (reiterated)
- Grammar is productive; all languages change with time; lexical growth (e.g., “googling”) and broader changes in phonology, syntax, etc.
1.3 Studying Language Scientifically
- What it means that linguistics is a science
- Not about lab coats; about systematic, empirical observations to describe phenomena
- Aim: objective, descriptive observations, avoiding value judgments
- Viewing language scientifically requires descriptive descriptions of actual language use
- Accessing the mind and metalinguistic awareness
- Much linguistic knowledge is unconscious; metalinguistic awareness is conscious knowledge about grammar
- School-taught language knowledge is often metalinguistic; first language acquisition involves implicit knowledge
- Examples of metalinguistic awareness:
- Creating new words (e.g., blifter vs lbitfer): both use English phonetics, but one form is not a possible English word; this supports a descriptive judgment about grammaticality
- Grammaticality vs acceptability judgments: some forms are grammatical in principle but there are acceptable and non-acceptable question forms
- Distinguishing grammatical vs ungrammatical forms is about what the mental grammar can generate, not about prescriptive norms
- Observing what’s possible in a language
- Two similar sentences can both be acceptable in declarative form and some question forms, but wh-questions reveal constraints (e.g., what vs what with a dangling piece)
- (e) is grammatical; (f) is not; the difference is explained by the mental grammar
- Tools and methods for language science (introductory overview)
- Perceptual/experimental methods to access metalinguistic judgments: acceptability judgments, surveys, interviews
- Corpora: large databases of real language use (written, spoken, sign language video, transcripts)
- Accessibility and practicality: these tools are relatively easy to use and accessible to learners
- Specialized tools used by phoneticians and sign-language researchers:
- Praat for phonetic/phonological analysis of audio
- ELAN for video annotation
- SLP-Annotator for sign language phonetics
- Additional measurement techniques from behavioral psychology and neuroscience:
- Reaction times, reading times, listening tasks
- Eye-tracking during reading/listening/watching sign language
- EEG and fMRI to observe brain activity during language processing
- The meta-skill of doing language science
- Even if you’re not wearing a lab coat, you can make systematic observations and interpret them to infer about the mind
- Practical examples of data collection
- Acceptability judgments across speakers, surveys to map regional variation, corpus analyses for word usage and phrasing
- Note on limitations and biases
- Observations reflect the observer’s methods and communities; ethical and social biases can influence research outcomes
1.4 Thinking About Standards and “Proper” Grammar
- Reiterates the radical goal: treat all languages and dialects as equally valid from a linguistic perspective
- Language and power dynamics are intertwined
- Language attitudes and expectations are connected to social hierarchies (teacher-student, doctor-patient, customer-service, etc.)
- People may internalize biases about which varieties are “standard” or “proper”
- Language standards are not neutral; they are tied to economic, social, or political power
- Examples and discussion of standard language concepts
- Standard forms often align with the language used by educated, typically white, members of society
- In practice, standard varieties are enforced via norms, dictionaries, style guides, textbooks, and spell-checkers rather than official police
- The term “standard” often reflects the language used in capitals or by dominant political groups; in many cases it aligns with White, educated speech
- Case study: Black English varieties (AAL/AAVE) as markers of Black identity; not “standard” in the sense of UK/Queen’s English; prejudice and misperceptions around standardization
- Consequences of assuming a single standard
- Non-standard language varieties can be unfairly judged as inferior in education and professional contexts
- Several real-world examples illustrate how biases affect education, legal judgments, and perceptions of competence
- Encouragement to develop metalinguistic awareness of biases and standards while recognizing the grammar itself is equally valid across varieties
- Note on terminology and future chapters
- Chapter 2 will discuss terms like language, variety, and dialect in more detail
1.5 Doing Harm with Language Science
- Language science has a history of harm, often tied to colonialism and missionary work
- Early linguistics documented Indigenous languages to aid conversion and colonization, using European writing systems that distorted phonetics
- Writing froze certain language forms and could erase variation; missionary writings sometimes claimed Indigenous languages were inferior
- The Canadian residential school system forcibly separated Indigenous children from their families and languages; many languages were suppressed or lost as a result
- Dual nature of documentation
- While some documentation has helped revival efforts (e.g., Huron-Wendat), it also caused harm and misrepresentation
- The field has also historically been linked to assimilation policies and linguistic discrimination
- Modern ethical concerns in language science
- Data collection can exploit language communities; researchers must ensure benefits to communities and avoid extractive practices
- Data sharing and publication must respect privacy and sacred knowledge
- Risk of linguistic and cultural appropriation when researchers speak for a community without its members
- Descriptive work can inadvertently lead to prescriptive norms if researchers’ observations become de facto standards
- The need for humility and reflexivity
- Science is one way of knowing with its own biases and limits
- Move from viewing grammar as prescriptive rules to seeing it as a living, shared mental system
- Recognize that language exists within communities and is co-constructed through social interaction
- Preview: later chapters will address more examples of harm and bias
1.6 Doing Good with Language Science
- Emphasizes constructive applications of language science
- In tech: language-trained professionals improve software that summarizes, translates, and synthesizes speech; voice assistants and GPS systems
- L1/L2 learning apps use metalinguistic awareness to aid language learning
- Revitalization and reclamation: linguistic analysis supports Indigenous language teaching materials and adult language learning,
with emphasis on community leadership rather than researcher-driven “savior” narratives
- Diverse career applications for linguists
- ESL teaching certification and language teaching more broadly
- Speech-language pathology: clinical applications for brain injury and language impairment
- Accent/dialect coaching for media and entertainment
- Consulting roles: contract interpretation, brand-name development, forensics (authorship), etc.
- Language is pervasive and central to human interaction
- Understanding language supports people and can contribute to social good
1.7 Do Chatbots Have Language?
- Distinguishing human language from AI-generated text
- Human language is generative and governed by systematic principles; shared across people via mental grammar
- Generative AI (LLMs) produce text by statistical patterns learned from large data; they do not have true semantics or understanding
- What LLMs do
- Trained on vast datasets, often without explicit consent from creators; training data largely consists of copied or repurposed material
- LLMs do not have meaning; they predict likely word sequences based on frequency patterns
- Output only has meaning when humans interpret it; it lacks intrinsic understanding
- The limitation of LLMs in truth-telling
- Researchers argue LLMs struggle with truthfulness; they optimize for plausible output, not factual accuracy
- The debate is about what counts as “generative” and what counts as “intelligent” in AI
- Key takeaway
- Human language is generative and grounded in a mental grammar; AI text generation is combinatorial and statistic-based without semantics
- Reference point provided: critique of AI’s truthfulness and the nature of AI-generated language
1.8 Exercise Your Linguistics Skills
- Exercise 1: First language (L1) observations
- Identify your L1(s); make two scientific observations about your L1 (descriptive, not prescriptive)
- Exercise 2: Naming a new product
- Propose a unique product name; evaluate other students’ proposed names for marketability and linguistic fit
- Exercise 3: Create new English verbs from nouns
- Demonstrate productivity of mental grammar by coining three new verbs from existing nouns; illustrate with sentences
- Exercise 4: Document a new word’s usage
- Observe a recently entered English word in context; draft a dictionary-style definition based on usage
Key Terms and Concepts (glossed)
- Descriptive vs Prescriptive: Descriptive describes actual language use; prescriptive prescribes rules
- Mental grammar: the internal knowledge of language in the mind, including phonology, morphology, syntax, semantics, pragmatics
- Modality: vocal (speech) vs signed (sign language) language
- Lexicon: mental dictionary linking form (sound/sign/written form) to meaning
- Phonetics/Phonology: articulatory phonetics; how sounds are organized and function in a language
- Morphology: structure of words and internal meaningful pieces
- Syntax: rules for combining words into phrases and sentences
- Semantics: meaning of words and sentences
- Pragmatics: how meaning depends on context
- Acceptability judgments: judgments about what is possible or acceptable in a language, used to access the mental grammar
- Corpora: large databases of natural language usage
- Praat: software for phonetic analysis of audio
- ELAN: software for annotating video data
- SLP-Annotator: tool for sign language annotation
- Metalinguistic awareness: conscious knowledge about language
- Standard language: prestige-variety often tied to power; not linguistically superior
- Language harm and ethics: colonialism, residential schools, data misuse, cultural appropriation
- Generative AI vs human language: LLMs vs mental grammar; semantics and truthfulness differences
ext{Plural formation (illustrative, descriptive):} ext{regular plural}
ightarrow ext{cookie}
ightarrow ext{cookies} \
ext{Note: this is the common pattern described in the text; irregularities exist in many languages}
- Engaging with the material ethically and practically:
- Use descriptive observations when studying language use
- Be mindful of power dynamics and cultural context when researching or describing languages
- Consider how technology (AI) interacts with language and its study, including issues of meaning, accuracy, and bias