Counting English words, from 520,779 entries to daily use
· 10 min read
The OED lists 520,779 entries (Jan 2026) but only ~170K in current use. Why dictionary counts mislead and what vocabulary size matters for writers.
TL;DR The Oxford English Dictionary lists 520,779 entries (Jan 2026), but only ~171,476 are in current use. The gap comes from obsolete words, derivative forms, and methodological differences between dictionaries. English adds ~1,000–2,000 words yearly but loses hundreds to obsolescence.
The short answer: the Oxford English Dictionary contains 520,779 entries as of January 2026, alongside 888,251 meanings and 3.9 million quotations, with about 1,000–2,000 new words and senses added annually (Wikipedia's OED entry-count survey). But the number you get depends entirely on what you count — and why.
Google "how many words in the English language" and you'll see figures ranging from 170,000 to over a million. The gap isn't an error. It's the difference between counting dictionary headwords versus words people actually use, between lemmatized entries and distinct word forms, between general vocabulary and specialized terminology. Understanding what each number measures tells you more about English than any single total ever could.
What Is the English Language Word Count?
There is no single authoritative count because "word" has no universal definition in linguistics. The Oxford English Dictionary's online edition contains 520,779 entries as of January 2026, making it the largest historical record of English (Wikipedia's OED entry-count survey). Merriam-Webster's Unabridged Dictionary lists about 470,000 entries focused on current usage (Merriam-Webster FAQ). Corpus linguistics — analyzing billions of words of actual text — puts the active vocabulary at roughly 171,476 words — a corpus-era estimate tied to the print edition, per the OED's recorded edition history (Wikipedia).
Key Takeaway: Dictionary entries ≠ living vocabulary. The OED's 520,779 includes 47,156 obsolete words and 9,500 derivative subentries. If you want words people actually use today, the corpus number (~171,476) is the better benchmark.
The discrepancy matters for practical reasons. A Scrabble player consults the Official Scrabble Players Dictionary (187,000 playable forms). A novelist checks a collegiate dictionary (165,000 current words). A historical linguist needs the OED (520,000+ historical forms). Each tool serves a different purpose, and using the wrong one leads to bad decisions — like rejecting a valid word in a game or misunderstanding a historical text.
If you're working with your own text, ToolSura's word counter helps you check word count, unique words, and frequency in your browser — no upload, no tracking.
Even within a single dictionary, the count shifts depending on what you include. The classic decomposition comes from the 1989 print edition, which the OED itself recorded as 291,500 total entries and 231,100 main entries, 47,100 of those already obsolete — figures preserved in the OED's own edition history (OED edition history, archived). The oft-quoted "171,476 words in current use" is a corpus-era estimate tied to that print edition, not a decomposition of today's 520,779 online entries, which mix in draft-revision material the print set never had. That's still more than many collegiate dictionaries because the OED preserves every word it can document, while collegiate dictionaries prune aggressively for relevance.
Quick Comparison: Dictionary Counts at a Glance
| Source | Entries | Scope | Best For |
|---|---|---|---|
| Oxford English Dictionary online (Jan 2026) | 520,779 | Historical + current | Academic research, etymology |
| Merriam-Webster Unabridged | ~470,000 | Current US usage | General reference |
| OED print edition (1989) | 291,500 entries; 231,100 main | Full recorded history | Etymology, historical work |
| Official Scrabble Dictionary | 187,000 | Game-legal forms | Competitive play |
Why the Number Keeps Changing
English grows by roughly 1,000–2,000 new words and senses per year, the online OED has long published quarterly updates (Wikipedia). The September 2024 update added 690 entries; June 2024 added ~1,100; March 2024 added ~700 (OED blog, archived). But hundreds of words also fall out of use annually — marked "obsolete" or "archaic" in the OED. Net growth is closer to 500–800 words per year.
This isn't new. The OED's 1989 Second Edition had 291,500 total entries, 231,100 of them main entries (OED edition history, archived). The online edition (revising since 2000) has nearly doubled that. The difference reflects both genuine expansion and broader inclusion criteria — regional dialects, slang, technical terms, and digital-age coinages that earlier editions excluded. Words like "selfie," "hashtag," and "blockchain" entered the OED in the last decade; and "floppy disk" and "VHS" now read as relics rather than tools.
The rate of change accelerated with the internet. Before 1990, new words entered print slowly through books and periodicals. Today, a term can go from Reddit thread to global usage in weeks. The OED's revision cycle — quarterly updates of ~1,000 entries — struggles to keep pace with digital vernacular. This lag means the dictionary always trails living usage by months or years.
Loanwords accelerate the growth further. English borrows freely — "karaoke" from Japanese, "schadenfreude" from German, "ballet" from French, "avatar" from Sanskrit. Each borrowing adds to the count, and English has absorbed words from a long roster of languages throughout its history. The OED's etymologies trace this web of borrowing, showing how English vocabulary grew through contact rather than invention alone.
How Dictionaries Count Words (and Why They Disagree)
Lexicographers follow specific but different protocols:
- Headword vs. form counting — The OED groups inflected forms under one headword ("run," "ran," "running" = one entry, three forms). Corpus studies typically count lemmas (the dictionary form) or word families ("run" + "runner" + "running" = one family). This alone can shift totals by 2–3x.
- Historical vs. current scope — The OED aims for historical coverage across its full range. Merriam-Webster prioritizes contemporary American English. A word last seen in 1450 stays in the OED; it disappears from collegiate dictionaries.
- Compound handling — English writes compounds as separate words ("ice cream"), hyphenated ("ice-cream"), or solid ("icecream"). Dictionaries often list all three as separate entries. German writes "Eiscreme" as one word — inflating English counts by comparison.
- Technical terminology inclusion — Medical vocabulary adds more than 3.5 million concepts; chemical nomenclature exceeds 10 million (NIH study). General dictionaries exclude most of these. Specialized dictionaries include them.
- Regional and dialect coverage — The OED now includes World Englishes (Indian, Singaporean, Nigerian, Caribbean). Earlier editions were British-centric. Each variety adds thousands of region-specific terms.
These methodological differences explain why you'll never get a straight answer. The count depends on whether you include obsolete forms, how you treat inflections, whether compounds count as one or many, and which Englishes you consider "standard."
Key Terms You'll Encounter
- Lemma (headword)
- The dictionary form of a word — the form under which it's listed (e.g., "run" for "run," "ran," "running").
- Word form
- Any distinct inflected or derived version of a lemma (e.g., "run," "ran," "running" are three forms).
- Word family
- A lemma plus its closely derived forms (e.g., "run," "runner," "running," "rerun").
- Headword
- The citation form at the top of a dictionary entry; often the same as the lemma.
- Obsolete word
- A word no longer in general use since a specific date (e.g., 1755), marked as such in the OED.
- Derivative subentry
- A word derived from a headword, listed under it rather than as its own main entry (e.g., "runner" under "run").
Active vs. Passive Vocabulary: What You Actually Use
The 20,000–35,000 figure you often see refers to receptive vocabulary — word families you recognize when reading or hearing (Nation 2006, Reading in a Foreign Language).
Citation: Nation, I. S. P. (2006). How large a vocabulary is needed for reading and listening? Reading in a Foreign Language, 18(1), 1–14.
Your productive vocabulary (words you actually use in speech and writing) is smaller: roughly 5,000–10,000 words for daily communication. The rest you understand but don't deploy.
This distinction matters for writers. If you're checking whether your audience will know a word, the receptive number applies. If you're measuring your own writing range, the productive number is the reality. ToolSura's word counter tool helps you see your actual productive vocabulary in any text — paste a draft and watch the unique word count.
For practical writing tasks, knowing how many words in a paragraph or essay word count guidelines helps you hit the right targets.
Vocabulary size also correlates with reading habits. Nation (2006) reports that encountering a word 10–12 times in varied contexts typically moves it from receptive to productive knowledge. That's why wide reading expands usable vocabulary faster than memorizing lists — context provides the repetition and variation that cements words in active memory.
For practical fluency, you need fewer words than you think. Nation's research found that ~3,000 word families covers 95% of everyday text; ~8,000–9,000 reaches 98% coverage. The remaining 2–5% consists of low-frequency, specialized, or domain-specific terms that you encounter rarely and can often infer from context.
How Vocabulary Sizes Compare
| Vocabulary Scope | Approximate Size | Source |
|---|---|---|
| OED, all entries (2026) | 520,779 | Oxford English Dictionary |
| Merriam-Webster Unabridged | ~470,000 | Merriam-Webster |
| Words in current use | ~171,476 | OED corpus data |
| Shakespeare's complete works | ~17,677 | Warren King, cited in Wikipedia |
| Educated native speaker (receptive) | 20,000–35,000 | Nation 2006 |
| Daily productive vocabulary | 5,000–10,000 | Nation 2006 |
| Fluent reading (95% coverage) | ~3,000 word families | Nation 2006 |
| Near-full comprehension (98%) | ~8,000–9,000 word families | Nation 2006 |
For practical word counting, use how to check word count in any document or compare word count vs character count for precise measurements.
Shakespeare, Science, and the Limits of Counting
Shakespeare did not invent 1,700 words. He recorded roughly 1,700 words first in surviving print (Warren King via Wikipedia). Many existed in spoken English decades earlier; the OED's ongoing antedating project regularly pushes first citations back. The "Shakespeare invented X words" claim confuses textual survival with linguistic innovation. We know this because the OED finds earlier examples in letters, diaries, and legal documents that predate the First Folio.
Shakespeare's complete works use ~31,534 unique word forms — a remarkable range, but within the receptive vocabulary of a modern educated speaker. The myth of his massive invention stems from his texts being the earliest surviving record for many words, not from him coining them.
English does not have "more words than any other language." By dictionary entries, yes (520K vs. French ~100K, Spanish ~93K). By morphemes (meaning-bearing units), Korean and Japanese exceed English. By compound formation, German and Turkish generate infinite words productively — "Donaudampfschifffahrtsgesellschaftskapitän" is one word in German, four in English. The comparison only works if you define "word" as "dictionary headword in a major historical dictionary."
Scientific terminology breaks every count. Chemical abstracts register over 10 million named compounds. Biological taxonomy adds millions more. These follow systematic naming rules, not lexical creativity. Including them makes "English vocabulary" a moving target with no ceiling. The International Union of Pure and Applied Chemistry (IUPAC) naming system can generate a unique name for every possible molecular structure — effectively infinite.
Myth buster: Dictionary size ≠ language richness. A language with 500,000 dictionary entries but rigid morphology may express less nuance than one with 50,000 roots and flexible compounding. Richness lives in what a language lets you say, not how many entries a dictionary prints.
Related Tools & Further Reading
- Count your own text — Use ToolSura's word counter tool to check word count, unique words, and frequency in your browser. No upload, no tracking.
- Visualize word patterns — The word cloud generator shows which words dominate your writing.
- Practical word count guides — How many words in a paragraph, essay word count guidelines, how to check word count in any document, word count vs character count, words to pages converter.
The Bottom Line
English has far more words than any one person needs or uses. When you ask how many words are in English, the OED's 520,779 entries document the full recorded history, but the roughly 170,000 words estimated to be in current use are what circulate today, and the ~3,000 word families that cover 95% of everyday text are what most readers actually rely on. The "right" number depends entirely on the question you're asking: a lexicographer counts one way, a corpus linguist another, and a writer simply needs the vocabulary their audience already owns.
One thing is certain, though. English keeps growing. Roughly 1,000–2,000 new words and senses enter the OED every year through quarterly updates, while hundreds more quietly fade into obsolescence. Whatever total you cite today will be slightly stale tomorrow, which is why serious sources always anchor their numbers to a date.
Written by
Harriet Cole
My whole subject is that a word count is not one number, and the definition used changes it by a large enough margin to matter when a limit applies. Splitting on whitespace counts space-separated tokens, which treats a hyphenated word as one and a number with a decimal point as one. Splitting on a word-boundary regular expression splits both, and splits contractions in ways nobody expects. Counting characters is simpler and more ambiguous still, since whether spaces count changes the total by a few percent and whether a character means a code point or a byte depends on the encoding.
Sentence counting is the least reliable of the set. Splitting on a period followed by a capital breaks on abbreviations and on decimal numbers, so 3.14 and e.g. each produce a sentence boundary that does not exist. Any figure produced this way carries an error rate worth knowing about.
Paragraph counting needs a rule about blank lines, because some formats separate paragraphs with a blank line and others do not. Reading time is the most quoted and least examined. It comes from a words-per-minute figure, that figure is a population average, and it assumes continuous focused reading.
Technical text read carefully is slower, and skimming is faster. I state the rate on every page rather than presenting the minutes as a measurement. I also cover what the count is used for, because a limit is enforced somewhere and knowing whether the limit counts markup changes the answer substantially. Where a limit applies, I would rather the tool counted with the same definition the rule was written against than quietly applying its own.