Original research · 2026-06-24
The Most Polysemous English Words (2026)
Polysemy is the property of a single word carrying many distinct but related meanings. The ranking below is generated server-side at request time by counting the senses Wiktionary records for each word. The words at the top are not rare or technical, they are the shortest, oldest, most-used words in the language.
Spotted an error on this page? Report it and we'll review it.
What is polysemy?
A word is polysemous when it has multiple related senses under a single headword, as opposed to homonymy, where unrelated meanings happen to share a spelling (the "bank" of a river versus a "bank" that holds money). The verb draw is a textbook case of polysemy: to draw a picture, to draw a card, to draw water from a well, to draw a crowd, to draw a conclusion, and to draw a match all share a common metaphorical core, the idea of pulling something out or along, extended into dozens of contexts. Lexicographers split that core into separate numbered senses, and the count of those senses is what this page ranks.
Crucially, the sense count is a property of how a dictionary divides meaning, not a physical constant. Different dictionaries split senses at different granularities, the Oxford English Dictionary famously records over 400 senses for set, far more than the count here. The ranking below reflects Wiktionary's editorial granularity. What is robust across any dictionary is the relative ordering: the busiest words in English are reliably the most polysemous, whoever counts.
Top 15 English words by recorded sense count
Counts reflect distinct senses recorded in Wiktionary for common, lowercase headwords (proper nouns excluded, see Methodology). Each word links to its full PlainSpell entry.
Frequency drives meaning
The single strongest predictor of how many meanings a word carries is how often it is used. This is not a coincidence of this corpus, it is a well-documented regularity sometimes called Zipf's law of meaning: across languages, a word's frequency and its number of senses rise together, roughly as a power law. The mechanism is intuitive. Every time a community reaches for a familiar word to name a new situation, the word stretches to cover it, and the new use, if it sticks, becomes another sense. Words that are reached for constantly stretch the most. Rare words, by contrast, are typically rare because they are precise, they name one thing and are retired when that thing is not in view.
Light verbs and the phrasal-verb explosion
Notice how many of the top entries are verbs: draw, take, go, run, break, set, strike, make. English leans heavily on a small set of semantically "light" verbs that contribute little fixed meaning of their own and instead take on meaning from their objects and particles. Take illustrates this vividly: take a photo, take a bus, take offence, take a wife, take ten minutes, plus the phrasal family take off, take on, take in, take up, take over, take after. Each particle combination is a candidate for its own dictionary sense. Because English is unusually rich in phrasal verbs, its light verbs accumulate senses faster than the same concepts do in languages that prefer separate prefixed verbs (German aufnehmen / annehmen / einnehmen distribute across distinct headwords what English packs into take). The phrasal-verb system is, in effect, a meaning multiplier bolted onto a handful of core verbs.
Why short, Anglo-Saxon words dominate
Almost every word near the top is a short word of Old English (Germanic) origin rather than a longer Latin or Greek borrowing. There are two reasons. First, the native Germanic core of English supplies the everyday, high-frequency vocabulary, and frequency drives polysemy as above. Second, borrowings tend to enter the language to fill a specific gap and carry a narrower, more technical remit, "comprehend" arrived to do a particular job, while the native "get" sprawled across get a job, get the joke, get home, get sick, get going. The result is a vocabulary stratified by polysemy: a small set of ancient monosyllables doing enormous semantic work, surrounded by a vast periphery of specialised loanwords each minding a single meaning.
Why this matters for readers, writers, and language tools
Polysemy is the hidden tax on plain language. The words that look simplest, the short familiar ones, are often the most ambiguous, because a reader must select the intended sense from dozens on the fly. "The board will draw on reserves to set the rate before the meeting runs over" is built almost entirely from top-15 words, and every one of them is doing context-dependent work. For language-model and search systems, these high-polysemy words are where word-sense disambiguation is hardest and most consequential, getting "run" wrong (a jog, a tear in a stocking, a sequence, an election bid, the operation of a machine) cascades into a wrong interpretation of the whole sentence. For writers aiming at clarity, the practical lesson is the inverse of the usual advice: a precise low-frequency word sometimes carries less ambiguity than the short, friendly, wildly polysemous one it replaced.
Methodology
For each English headword, PlainSpell stores the full set of Wiktionary sense definitions. The sense count is simply how many distinct senses Wiktionary records for that word. The ranking reflects the current dataset, restricted to lowercase, alphabetic-only headwords that carry a frequency rank of 50,000 or better, ordered by sense count.
Proper-noun filter (important): the raw, unfiltered sense count is dominated by proper nouns, because Wiktionary records every distinct place, person, or product that shares a name as a separate sense (dozens of towns named "Springfield", for instance). Those are not cases of semantic polysemy, so this ranking restricts to lowercase, alphabetic-only headwords that carry a frequency rank, leaving genuine common-word polysemy.
Limitations: sense counts reflect Wiktionary's editorial granularity, not an absolute measure of meaning, a dictionary that splits or lumps senses differently would produce different numbers (the OED records far more senses for several of these words). The figures are comparable to one another within this single source but should not be read as canonical meaning counts. The frequency cut-off (rank ≤ 50,000) keeps the comparison to words common enough that their polysemy reflects everyday usage rather than archaic or specialist sense inflation.
Sources
Source: Wiktionary (English edition) JSONL dump via wiktextract · 2026 Open data under CC BY-SA 4.0. Sense definitions per headword.
Source: Zipf, G.K., The Meaning-Frequency Relationship of Words The Journal of General Psychology 33(2) · 1945 Original formulation of the frequency–polysemy power law.