← all systems · see it in The System
Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon
A word is a basic element of language that carries meaning, can be used on its own, and is uninterruptible.
Properties published scienceour proposal
Sourced fields, and our own readings.
| born at | with language (rung 11); 'neologism' first attested in English in 1772; word classes first systematised by Dionysius Thrax, 1st century BC |
| born precision | single uses to generations (a word); centuries (the study) |
| boundary | Inside: the word as linguists describe it (what counts as a word across languages: phonological, grammatical, orthographic and semantic criteria; lexemes and lemmas; word classes; word edges in speech and writing), the mental lexicon that holds a person's words (forms, lemmas, meanings, their links, frequency and neighbourhood effects), how words are learned (segmentation, early word meanings, fast and slow consolidation) and retrieved (the stages of saying a word, the tip-of-the-tongue state), and how new words arise, spread, change and are lost in a community (coinage, diffusion, regularisation, replacement, frequency laws). Outside: morphology as a field (roster sibling sys:morphology-linguistics), the sound system (sys:phonology), syntax and sentences, reading as a skill (sys:reading) and writing systems (sys:writing-system), booked as partners. |
| nested level | Language's fourth floor (sys:language L4: words and grammar, the lexicon). The word sits between the sound system it is built from (phonology) and the sentences it is combined into; it is held in each speaker's brain (runs on sys:human-brain at L6) and shared in a community (runs on sys:human-society at L7). |
| scale | social |
Measures (28) published science
| vocabulary of an average 20-year-old American English native speaker | 42,000 (5th-95th percentile 27,000-52,000) lemmas |
| word families behind that vocabulary | 11,100 word families |
| adult vocabulary growth, age 20 to 60 | about one new lemma every 2 days (6,000 lemmas in 40 years) lemmas per day |
| words spoken per person | about 16,000 word tokens per day |
| exposure needed for infants to tell words from part-words | 2 minutes |
| age by which infants know common nouns | 6 months |
| picture naming: onset of articulation | about 600 ms |
| picture naming stages: concept / lemma / form / syllabification / phonetic encoding | 175 / 75 / 80 / 125 (25 per phoneme) / 145 ms |
| settling time before a new word competes with known words | overnight sleep (tests at 0, 12, 24 h); a week after one 36-presentation session hours to days |
| English lexicon in print | 544,000 (1900); 597,000 (1950); 1,022,000 (2000) words |
| English lexicon growth | about 8,500 words a year; over 70% in 50 years words per year |
| English lexicon undocumented in standard references | 52 % |
| Old English irregular verbs still irregular | 177 tracked; 145 in Middle English; 98 today verbs |
| half-life of an irregular verb | scales as the square root of usage frequency (100x rarer regularises 10x faster) scaling |
| share of variation in lexical replacement rates explained by use frequency | approximately 50 % |
| Indo-European comparison | 200 meanings in 87 languages meanings x languages |
| languages sampled for sound-meaning associations | nearly two-thirds of the world's languages; 100 basic vocabulary items share |
| Zipf-Mandelbrot law parameters | alpha about 1, beta about 2.7 exponent, shift |
| languages where length tracks information content better than frequency | 10 languages |
| semantic model basis | 985 common English words; 7 subjects, more than 2 hours of stories |
| neologisms tracked on Twitter | 99 words |
| dogs that learned many object names quickly | 7 of 34 dogs |
| exposures a gifted dog needed for a new name | 4 exposures |
| ASL signs in ASL-LEX 2.0 | 2,723 signs |
| word classes in Dionysius Thrax's Greek grammar | 8 classes |
| word spacing in Western writing | about a thousand years old |
| criteria for 'word' tested cross-linguistically, none necessary and sufficient | 10 criteria |
| first English attestation of 'neologism' | 1772 year |
Open contentions (8) open contention
Where sources or readings disagree. Both stay on the map.
- C-413 Where do thoughts come from?
booked: No theory is proven, so the map lists the main ones side by side and doesn't pick one. It shows where thinking is handled in the brain, not where thoughts start.
the other reading: - C-2747 Is the link between a word's sound and its meaning arbitrary?
booked: The map keeps the competing readings; none is ruled.
the other reading: arbitrary by voluntary imposition (Locke 1690, as the encyclopedia quotes); non-arbitrary associations widespread in 100 basic items (Blasi et al. 2016); iconicity a pervasive trait of sign language (Sehyr et al. 2021) - C-2773 How do babies first learn what words mean?
booked: The map keeps the competing readings; none is ruled.
the other reading: referential intentions grasped by 6 months; words learned before intention reading - C-2781 What is the mental lexicon: a store, a network, or no store at all?
booked: The map keeps the competing readings; none is ruled.
the other reading: semantic network theory; spectrum theory; dual-coding theory; Chomsky's nativist view; connectionist emergent entries; hybrid models; Elman: no mental lexicon; words as stimuli that alter mental states - C-2796 Is word meaning held in the left brain, or both sides?
booked: The map keeps the competing readings; none is ruled.
the other reading: left-lateralised (lesion studies, as Huth et al. report); bilateral (Huth et al. 2016) - C-2800 What causes the tip-of-the-tongue state?
booked: The map keeps the competing readings; none is ruled.
the other reading: transmission-deficit model (Burke and colleagues); blocking/interference model; metacognitive monitoring model - C-2802 Is 'word' a universal unit across languages?
booked: The map keeps the competing readings; none is ruled.
the other reading: consistent definitions at different levels of description (Dixon & Aikhenvald 2002, as the encyclopedia reports); no good basis for a general, cross-linguistically viable word concept (Haspelmath 2011); avoid the term and work with morphemes (Harris 1946, as the encyclopedia reports); word as an everyday convention (as the encyclopedia reports) - C-2803 Why do word frequencies follow Zipf's law?
booked: The map keeps the competing readings; none is ruled.
the other reading: random or concatenative processes; preferential reuse; communicative optimisation; semantic organisation; mixtures and multiplicative processes
Cross-scale patterns seen here (19) our proposal
Shapes that recur when the same kind of problem is solved at different scales, and how each shows in Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon. Each links to every other system that shows it.
- borrowed-then-held (structural)
A word is first supplied from outside, named by others in daily life (infants know common nouns from 6 months; gifted dogs learn names in play), then held in the learner's own lexicon and retrieved without the outside cue; held only after consolidation in sleep. - nested ladder (structural)
A new word climbs a ladder of acceptance where each rung implies the ones below: nonce word (used once), protologism (a small group), prelogism (gaining use), neologism (accepted by institutions). - same role, different builds (structural)
The job of marking where one word ends and the next begins is done by different structures: syllable statistics, fixed stress, vowel harmony (Turkish), a two-syllable minimum (Walmatjari), spaces in alphabetic writing, kanji-kana switching in Japanese. - same toolkit, independent builds (structural)
From the same toolkit of speech sounds, unrelated languages across continents build the same sound-meaning links for basic words ('tongue' with l, 'nose' with n, 'small' with i), often independently rather than inherited or borrowed. - seed (structural)
A finite store of word families (11,100 behind 42,000 lemmas) combined with morphology unpacks a word supply with no upper bound: the number of word types is boundless because speakers keep coining, and words combine into phrases and sentences. - templating (structural)
A word is a module reused unchanged across sentences and moved between languages where it still works: typhoon (from Chinese, 1588), origami (Japanese, 1880), robot (Czech, 1921). - the whole sets the local reference (structural)
What a word is and what it means are read against the whole language: 'word' can be defined only language by language (Haspelmath); meaning is use in the language (Wittgenstein); a word's semantic position is its co-occurrence with 985 others across a corpus (Huth). - who opens the door (structural)
At the crossing from sound to the lexicon the receiver opens the door ahead of time: a heard form is a word only if the listener holds an entry for it; cortical responses to pseudowords become word-like only after training followed by overnight consolidation (a port built by learning). - coordination-before-mapping (partial)
Learners passing words to learners converge on a structured, easier-to-learn lexicon with no designer and no plan of the whole. - namespace completeness (the map predicts its gaps) (partial)
The lexicon's structure predicts words no list holds: knowing one member of a word family helps understand the others; 52% of English words in books are in no standard dictionary, found by counting rather than listing. - outside-to-inside (partial)
The partition of speech into words is first a pattern in the input (syllable statistics in a continuous stream; spaces in writing) and becomes the learner's own word forms. - partner-becomes-part (partial)
Two words become one: compounds (typewriter), contractions (can't), blends (brunch from breakfast and lunch), and the fused word is passed on as one. - symmetry breaking (name the cut) (partial)
Before convention any sound could name a meaning (Locke: chosen 'by a voluntary imposition'); a community's choice breaks that symmetry, and the chosen form then anchors a whole word family. Blasi et al. show the starting symmetry is itself tilted by sound-meaning biases. - The carry (partial)
The wait: a new word held fast by the hippocampus cannot settle into the lexicon (compete with its neighbours) until overnight consolidation has run; after one intensive session the effect appeared a week later. - The Payment to Entropy (partial)
Every hand-off of a new word pays a toll: most lexical innovations soon fall into disuse; in retrieval, transmission between meaning and sound can momentarily fail (the tip-of-the-tongue state). Form: signal and attention; no per-hand-off efficiency given. - trade one property for another (partial)
Word length is traded against predictability: across 10 languages length tracks average information content, so predictable words are short and less predictable ones long. - Umwelt: each system carves its own slice of one shared medium (partial)
Each language carves its own word classes and word edges from one shared world: English has articles and Japanese does not; Lushootseed treats noun-like words as predicates; what counts as a word differs by language. - not-diluting wins (surface)
Frequent irregular forms hold while rare ones regularise; over 1,200 years the -ed rule took over as irregulars fell from 177 to 98. The source reads this as rules competing, so the dark twin (winning by competition) is close. - same shape, opposite sign (surface)
Sound-alike neighbours speed recognition in lexical decision (neighbourhood effect), while a newly consolidated neighbour slows recognition of an existing word (lexical competition). The sources do not name one process at one level for both.
Valence: what it trades, and with whom (21) published science
Each exchange with a neighbour, read from both sides: who gains, who pays, and what carries it.
- Lexicon: the stock of words and word parts of a language, and as each person holds it (the mental lexicon), as linguistics and psychology describe it via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Syntax: how a language combines words into phrases and sentences (constituency and dependency, word order, agreement, recursion), how speakers build and parse it, how children learn it, how it differs across languages, and the field that studies it via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Literacy via sys:written-record · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Morphology (linguistics): how a language builds its words from smaller meaningful parts (roots, affixes, compounding, inflection and derivation), how languages differ in this, how speakers learn and use it, and the field that studies it via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Human (Homo sapiens) via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Language (human language) via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Sign language (the natural signed languages) via sys:electromagnetic-field · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Human brain via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Human society (from local community up to societies) via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Hippocampus (memory) via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Phonology: how languages organise the meaningless units their words are built from (phonemes in spoken languages; handshape, location and movement in signed ones), together with the field of linguistics that studies those systems via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Writing system (a script plus the rules binding it to a language) via sys:written-record · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Social network via sys:written-record · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Sleep via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Reading via sys:written-record · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Memory via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Speech (human spoken language) via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Dog via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- English language: a West Germanic language of the Indo-European family, from its Old English beginnings to its varieties and its use as a global lingua franca, as linguistics and its history describe it via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Semantics: meaning in language (the meanings of words and sentences, sense and reference, how meanings combine, how they are held in minds and brains and shared in a community) and the field of linguistics and philosophy that studies it via sys:acoustic-wave · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
- Brand via sys:mental-representation · Word: the unit of a language's lexicon (spoken, signed or written) as linguists describe it, and as it is held, learned and retrieved in each speaker's mental lexicon gain
Partners and parts published science
Built from / builds (12)
- Lexicon: the stock of words and word parts of a language, and as each person holds it (the mental lexicon), as linguistics and psychology describe it
- Syntax: how a language combines words into phrases and sentences (constituency and dependency, word order, agreement, recursion), how speakers build and parse it, how children learn it, how it differs across languages, and the field that studies it
- Concept
- Morpheme
- Phonology: how languages organise the meaningless units their words are built from (phonemes in spoken languages; handshape, location and movement in signed ones), together with the field of linguistics that studies those systems
- Lemma
- English language: a West Germanic language of the Indo-European family, from its Old English beginnings to its varieties and its use as a global lingua franca, as linguistics and its history describe it
- Competition between an established sense and an emerging sense for a word form's use
- Competition among word candidates in recognition and production
- Replacement of a meaning's word by an unrelated word across related languages
- Competition between near-synonyms for a topic's use
- A related name blocking the intended word in retrieval (one hypothesis of the tip-of-the-tongue state)
Sits inside / holds (2)
expressed by (10)
- Generation/Synthesis
- Recombination/Composition
- Templating / Standardization
- Interoperability/Transduction (Bridging)
- Encapsulation & Modularity (Abstraction)
- Sensing/Scanning
- Uncertainty Reduction/Learning
- Social Learning / Public Information Use
- Self-Monitoring
- Quality Control & Error Correction
floor filled by (25)
- Lexicon: the stock of words and word parts of a language, and as each person holds it (the mental lexicon), as linguistics and psychology describe it
- Syntax: how a language combines words into phrases and sentences (constituency and dependency, word order, agreement, recursion), how speakers build and parse it, how children learn it, how it differs across languages, and the field that studies it
- Electromagnetic field
- Acoustic wave (sound: pressure waves in matter)
- Language (human language)
- Written record (durable external marks: texts, archives, canon)
- Human brain
- Human society (from local community up to societies)
- Concept
- Hippocampus (memory)
- Morpheme
- Phonology: how languages organise the meaningless units their words are built from (phonemes in spoken languages; handshape, location and movement in signed ones), together with the field of linguistics that studies those systems
- Writing system (a script plus the rules binding it to a language)
- Social network
- Sleep
- Reading
- Memory
- Statistical learning
- Morphology (linguistics): how a language builds its words from smaller meaningful parts (roots, affixes, compounding, inflection and derivation), how languages differ in this, how speakers learn and use it, and the field that studies it
- Lemma
- Mental lexicon
- Speech (human spoken language)
- Dictionary
- English language: a West Germanic language of the Indo-European family, from its Old English beginnings to its varieties and its use as a global lingua franca, as linguistics and its history describe it
- Semantics: meaning in language (the meanings of words and sentences, sense and reference, how meanings combine, how they are held in minds and brains and shared in a community) and the field of linguistics and philosophy that studies it
Kinds (11)
- Compound word
- Contraction
- Nonce word
- Protologism
- Neologism
- Loanword
- Blend word
- Sign (lexical item of a sign language)
- Phonological word
- Grammatical word
- Orthographic word
wired (6)
- Morpheme
- Phonology: how languages organise the meaningless units their words are built from (phonemes in spoken languages; handshape, location and movement in signed ones), together with the field of linguistics that studies those systems
- Writing system (a script plus the rules binding it to a language)
- Grammar (the structure of a language, as linguistics describes it)
- Merge (linguistics)
- Links between words by sound and meaning
Notes published science
- Adults never stop adding words: an average American English speaker knows about 42,000 lemmas at 20 and about 48,200 at 60, roughly one new word every two days for forty years (src:brysbaert-2016).
- A word you have just learned does not yet compete with the words you already know. It starts to (slowing your recognition of look-alikes such as 'cathedral' after learning 'cathedruke') only after a night's sleep (src:davis-gaskell-2009).
- Across nearly two-thirds of the world's languages, unrelated languages keep choosing the same sounds for the same basic things: 'tongue' with l, 'nose' with n, 'small' with i. These links mostly arise independently, not by inheritance or borrowing (src:blasi-2016, abstract).
Layers
The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.
the boardroom (shared floors)
- L3: Word forms: the heard, read or signed form matched against stored forms; frequent words and words with many neighbours are recognised faster; Wernicke's area as one store of word forms for hearing and speaking
- L4: The lemma and its morphology: the word's grammatical identity (word class, gender) and the morphemes it is built from; one lexeme groups all its inflected forms
- L5: Meaning: the lexical concept, mapped in intricate patterns across both hemispheres' cortex and consistent across people; meaning read from use
- L6: The mental lexicon held: a person's vocabulary (about 42,000 lemmas at 20), linked by sound and meaning, learned in two stages (hippocampus fast, neocortex slow, consolidated in sleep) and retrieved on demand
- L7: The community's lexicon: words shared, coined, spread, regularised, replaced and lost across generations (about a million English word types in print, growing ~8,500 a year)
way in: Hearing: spoken words heard
- L1: Speech in the air from the people around: continuous, with no pauses between most words (about 16,000 word tokens spoken per person per day)
- L2: Segmentation: the listener cuts the stream into word-like units by syllable statistics (transitional probabilities) and by cues such as fixed stress
way in: One's own words heard and monitored
- L1: The speaker's own words, aloud or in inner speech
- L2: A self-monitoring network (cingulate, insula, SMA, motor areas, cerebellum, thalamus, basal ganglia) checks what is said
way in: Sight: signed words seen
- L1: Signs in light: each sign a lexical item with its own handshape, location, movement, frequency and iconicity (ASL-LEX lists 2,723)
- L2: Sign forms taken in with their phonological structure (handshapes of both hands, location, contact) and their iconicity
way in: Sight: written words read
- L1: Marks on a surface, with or without spaces between words (spaces in Western writing are about 1,000 years old)
way in: The named world: what a word is used about
- L1: Things in view while someone names one of them (a parent naming a picture)
- L2: The eyes go to the named thing
way in: Thought: a message to put into words
- L1: A preverbal conceptual message to express; where it comes from is contested (C-413)
way in: Use across the community: how often words are used
- L1: Words in use across a community's talk and books (about 1,022,000 English word types in print in 2000)
- L2: Use frequency: how often each word is used, near-Zipfian in every language
way out: Coining a word: a new word put into use
- L1: The new word used: nonce word, then protologism, prelogism, neologism
- L4: Built from existing parts: compounds, affixes, blends, acronyms, names
- L6: Speakers' boundless coining from what they know
- L7: A community with a concept that lacks a term (or a term that lacks detail)
way out: Saying a word: from concept to articulation (about 600 ms in picture naming)
- L1: The spoken word in the air (articulation starts about 600 ms after a picture appears)
- L2: Syllabification (about 125 ms, 25 ms per phoneme) and phonetic encoding (about 145 ms)
- L3: Phonological code retrieval (about 80 ms): the sound form fetched only after the lemma; where it fails, the tip-of-the-tongue state
- L4: Lemma retrieval (about 75 ms): related lemmas are active too and one is chosen
- L5: Conceptual preparation: the target concept selected (about 175 ms)
way out: Signing a word
- L1: The sign made in light: handshapes, location, movement, contact
- L6: A sign retrieved from the signer's lexicon
way out: Writing a word: the lexicon put on a surface
- L1: Marks on a surface, words separated by spaces and punctuation in modern alphabetic writing
- L6: The word as the writer's orthography sets it (one word, two words, hyphenated)
word: from the signal to the community's lexicon (reader's view)
- L1: The signal: words in the air, in light (signs) or on a surface (writing), mostly without gaps between them in speech
- L2: Segmentation: the stream cut into word-like units by syllable statistics, fixed stress, vowel harmony or (in writing) spaces
- L3: The word form: the phonological (or orthographic, or sign) word, with its own size rules, matched against stored forms
- L4: The lemma and morphology: grammatical properties (word class, gender) and the morphemes the word is built from; the lexeme groups its inflected forms
- L5: Meaning: the lexical concept, held in semantic maps that tile both hemispheres
- L6: The lexicon a person holds: about 42,000 lemmas at 20, one more every 2 days to 60, linked by sound and meaning, learned fast then consolidated in sleep, and retrieved on demand
- L7: The community's lexicon: words shared, coined, spread, regularised and replaced across generations