Research

Prior Art: What Has Been Tried Before

TL;DR

• World English is not the first attempt to make English easier. Roughly a century of controlled-language subsets, spelling and grammar reforms, and empirical intelligibility research sits behind it, and the pattern in that history is sharp: subsets that live inside a bounded domain survive; top-down reforms that ask everyone to respell or re-learn fail. That is direct evidence for the project’s own design principles — subtract before you add, stay legible — and for O5’s own local stopping criterion, adopt an existing informal spelling, never coin a new one (see the honest exceptions to that criterion in §B below).
• The one place the evidence contradicts World English is pronunciation: Jennifer Jenkins’ empirically-derived Lingua Franca Core finds the th sounds /θ/ and /ð/ are not needed for international intelligibility and are safely substitutable — yet pronunciation keeps them. That tension is recorded honestly in §C below.
• The one place it is strongly corroborated is grammar: English-based creoles and adult learner interlanguage independently converge on several of the exact moves World English makes — no third-person -s, preverbal negation, invariant tags, regularized past tense — suggesting these are what English becomes under simplification pressure, not arbitrary inventions.

How to read this file. Each entry follows one template: what it was · what it tried · what happened · lesson for World English · which WoE rules it bears on. Dates and claims were web-verified; where a source is weak or a figure varies, it is flagged inline. Sources are listed per-entry and consolidated at the end, in the style of PAIN-POINTS.

A. Subsets that survived

The common thread: each is a subset of ordinary English (a restricted vocabulary and a tamed grammar), deployed in a bounded context (news, aircraft manuals, business), with institutional backing. None asked anyone to spell differently. That is exactly the “subtract, stay legible” bet World English makes.

Basic English

• What it was. An 850-word regularized subset of English designed by C.K. (Charles Kay) Ogden, published in Basic English: A General Introduction with Rules and Grammar (1930), growing out of The Meaning of Meaning (Ogden & Richards, 1923).
• What it tried. One vocabulary (600 nouns, 150 adjectives, 100 “operations”/structure words) plus ~18 grammar rules, meant to serve at once as an international auxiliary, an on-ramp to full English, and a plain English — covering ~90% of everyday concepts.
• What happened. Briefly prominent (Churchill promoted it in 1943), then faded after WWII. It survives as a beginner word-list and an influence on VOA and Simplified Technical English.
• Lesson for WoE. A small core lexicon is powerful, but a word-list alone is not a language — Basic English under-specified grammar and over-trusted paraphrase. World English should pair its core word-list with the full rule specs, not lean on vocabulary limits alone.
• Bears on. The core lexicon; S3, S4.

VOA Special English (now “Learning English”)

• What it was. A controlled register broadcast by the Voice of America, first aired October 19, 1959 (Word Book first published 1962; rebranded “Learning English” in 2014).
• What it tried. World news in a ~1,500-word core vocabulary, short sentences, no idioms, read about one-third slower than standard broadcast English.
• What happened. Still produced daily — one of the most durable controlled-English projects, precisely because it is a delivery style, not a reform.
• Lesson for WoE. Slower, fully-articulated speech is a feature, not a defect — VOA proves learners rely on it and it scales. This is direct support for World English’s pronunciation stance.
• Bears on. P5 and P6 (careful, syllable-timed speech is always correct); the core lexicon.

ASD-STE100 Simplified Technical English (STE)

• What it was. A controlled natural language for technical documentation: ~65 writing rules plus a dictionary of approved / non-approved words. Developed in the late 1970s by AECMA (European aerospace industry) at airlines’ request; first issued as the AECMA Simplified English Guide in 1986.
• What it tried. Enforce one word = one meaning, one part of speech per word, short sentences, and active voice, so non-native mechanics could safely read maintenance manuals.
• What happened. Thriving and mandatory across aerospace/defense; now owned by ASD and maintained by the STE Maintenance Group (became an international specification in 2005; the organization states it became an international standard in 2025 — ⚠ vendor-stated, not independently confirmed).
• Lesson for WoE. The single closest precedent to World English’s method: a hard rule set plus a controlled lexicon, proven to work where safety and intelligibility matter most. “One word, one meaning” and “prefer active voice” are STE rules World English reaches independently.
• Bears on. S3 (one meaning), S2 (plain verbs), G9+S1 (prefer active), and the whole controlled-grammar approach.

Globish

• What it was. A ~1,500-word subset of English formalized and named (a global + English portmanteau) by Jean-Paul Nerrière in 2004 — a former IBM executive who had observed non-native speakers communicating with each other through the 1990s.
• What it tried. Codify the “common ground” non-natives already use for international business — explicitly “not a language” but a pragmatic tool for intelligibility, not nativeness.
• What happened. Persists as a commercial teaching brand (books, a foundation) rather than an institutional standard.
• Lesson for WoE. Intelligibility, not native-likeness, is the right target — Globish’s founding insight matches World English’s philosophy that ease and clarity outrank fluency-flavored polish.
• Bears on. The whole subset philosophy (README); the ~1.5–3k core-lexicon target.

Plain English / Plain Language movement

• What it was. A movement to make public, legal, and government writing clear. Key institutions: the UK Plain English Campaign (founded 1979 by Chrissie Maher) and the US Plain Writing Act of 2010 (signed October 13, 2010; US roots go back to 1970s paperwork-reduction efforts).
• What it tried. Eliminate jargon, gobbledygook, and officialese; require clear language the public can actually understand.
• What happened. Institutionalized and ongoing — mandatory for US federal agencies, influential in drafting standards worldwide.
• Lesson for WoE. Plain phrasing is adoptable at scale when it is guidance backed by authority, and it targets sentence- and document-level clarity — the exact scope of World English’s style spec.
• Bears on. The entire style spec (S1–S7).

B. Reforms that failed

The common thread: each asked the general public to spell or read differently, or existed only as a proposal. All failed or stalled. The only spelling changes in history that stuck are the conservative subset — which is precisely the line World English draws in O4/O5: adopt an already-common form, never coin a new one.

Noah Webster’s spelling reforms

• What it was. Americanized spellings promoted in Webster’s Compendious Dictionary (1806) and American Dictionary (1828).
• What it tried. A mix of conservative regularizations and radical phonetic respellings.
• What happened — the key data point. The conservative changes became standard American English: color (‹ colour), center (‹ centre), -ize, defense/offense, traveled (single -l-), plow, check. The radical ones all failed: tho, thru, tung, wimmen, fether, iland — Webster himself dropped several by 1828.
• Lesson for WoE — corrected. The sharpest lesson in the whole file: the same reformer’s conservative changes survived and radical ones died. World English’s O1 adopts exactly Webster’s winning set (color/center/-ize), and thru/tho are licensed by O5 precisely because they have since become common informal spellings — the condition Webster’s 1806 coinages lacked. But it is not accurate to say WoE “refuses his losing set except thru/tho.” WoE’s orthography reform is broader than “avoid Webster’s failures”: O2 drops silent etymological letters on its own criterion (worst offenders, kept to a short list), and that criterion produces several respellings — det, dout, receit, iland, suttle — that are original WoE coinages, not adopted informal spellings. iland in particular is a spelling Webster tried and dropped; WoE’s O2 revives it on different grounds (a purely silent letter), not because it has informal currency today. That is a conscious trade-off, not a coincidence: O2 accepts the risk Webster’s radical set ran into (a coined form with no existing usage to point to) for a small, curated list, betting that “short list, transparent rule, real silent letter” is enough to avoid Webster’s fate where “any letter that offends the eye” was not. The Cut Spelling comparison below answers with a related but distinct criterion — “keep the list small” — which is the same bet, not a different one, but neither O2 nor this file previously said so plainly; see “The coined-pronoun tension” below for the same trade-off applied to WoE’s pronoun paradigms.
• Bears on. O1 (the successful subset), O2 (the coinage exception, acknowledged), O4/O5 (the adopt-don’t-coin line, which O2 knowingly crosses for a short list).

Simplified Spelling Board (SSB)

• What it was. A US reform body announced 1906, funded by Andrew Carnegie (annual sums; the exact figure varies by source — ⚠ do not cite a single precise number). Members included Melvil Dewey and Mark Twain among the backers.
• What it tried. Promote simplified spellings, publishing a 300-word list in 1906 (over half of which were already common American forms).
• What happened. President Theodore Roosevelt ordered the Government Printing Office to adopt the list in August 1906; Congress reversed it in December 1906 and Roosevelt rescinded the order. The reform collapsed and never recovered.
• Lesson for WoE. Top-down, mandated respelling fails even with a president and a fortune behind it. This is the strongest argument for keeping World English’s orthography light, optional, and legible rather than a decreed rewrite — and for delivering predictability through the separate respelling key instead of changing the base spelling.
• Bears on. O4 (restraint); the whole orthography/pronunciation split.

Cut Spelling

• What it was. A reform scheme by Christopher Upward, published by the Simplified Spelling Society (handbook, 1991).
• What it tried. Delete redundant letters (not full phonetic spelling) — shorter words, better sound-symbol fit.
• What happened. No adoption; a niche enthusiast scheme.
• Lesson for WoE. Even modest, principled letter-deletion fails to spread when it touches ordinary spelling broadly. World English’s O2 does the same kind of thing (dropping silent b in debt → det) but confines it to a short list of the worst offenders, precisely to avoid Cut Spelling’s fate.
• Bears on. O2 (keep it small).

Initial Teaching Alphabet (ITA)

• What it was. A 44-character augmented alphabet devised by Sir James Pitman, introduced in English schools in 1961 as a transitional medium for teaching young children to read, before switching to standard spelling around age 7.
• What it tried. Make early reading easier with a near-phonemic symbol set, then hand off to normal orthography.
• What happened. Faded by the 1970s. The fatal flaw was transfer: children struggled to move from ITA back to standard spelling, and ITA reading material was scarce.
• Lesson for WoE. A phonetic learning layer must not replace the real orthography, or the transfer problem bites. World English’s respelling key is deliberately a parallel annotation over unchanged spelling — the opposite of ITA’s replace-then-transfer design — so there is nothing to transfer away from.
• Bears on. P1/P2 (respelling as a separate layer, not a new orthography).

Quirk’s Nuclear English

• What it was. A concept proposed by Randolph Quirk (~1982) for a restricted, culture-neutral “nuclear” core of English for international use. ⚠ Corroborated via the World Englishes literature (e.g. Close 1985), not Quirk’s own encyclopedia entry; do not confuse it with the unrelated later “Nuclear English” for the nuclear-power industry.
• What it tried. Strip English to an unambiguous common core suitable as an international medium.
• What happened. It stayed a proposal. It was never worked out into a full system or adopted — cited in the literature as a notable but unrealized idea.
• Lesson for WoE. The cautionary twin: a good idea for a reduced international English, never built, so it changed nothing. World English’s answer is methodology — write the specs, dogfood them (samples), and build the tooling — so it does not remain, like Nuclear English, a proposal.
• Bears on. The README methodology and the tooling backlog; the samples/regression discipline.

The coined-pronoun tension (Webster, again)

• The honest problem. The strongest lesson in this whole section is that reforms asking the public to re-learn something fail, and successes were subsets — “none asked anyone to spell differently.” World English’s grammar breaks that pattern in one place: G4’s pronoun paradigm (mes, hims, uss, yous, thems) and G12’s reflexives (meself, hisself) and G4’s plural-you possessive (you all’s) are not adopted from anywhere — they are coined forms a learner must memorize, in the exact structural position Webster’s tho/thru/tung/wimmen/ fether/iland occupied before most of them died. The creole-convergence evidence above defends WoE’s verb/tense moves (no 3sg -s, preverbal negation, invariant tags, regularized past) as natural endpoints of simplification, not designer inventions — but it says nothing about the pronoun coinages, because no creole or interlanguage source is cited for mes or hisself as a class. The “adopt an existing form, never coin” boundary (O5) is also applied selectively: it was used to reject a coined question particle (kept intonation instead, per P7 in §C), yet waived for the entire pronoun/reflexive system.
• Why WoE includes them anyway. Three differences from Webster’s failed coinages, stated plainly rather than left silent:
  1. Small and closed. Webster’s radical set was an open-ended, word-by-word rewrite of ordinary vocabulary (any word could get a new spelling); WoE’s coinages are a closed paradigm of a dozen or so pronoun/reflexive forms, learned once as a system rather than memorized word by word.
  2. Systematic and derivable. Every coined pronoun follows one visible pattern (object/ possessive form = subject form + -s; reflexive = possessive-looking form + -self), so once a learner sees the rule they can produce the rest — unlike Webster’s respellings, which had to be learned one word at a time with no generative rule linking them.
  3. Learnable in isolation, not competing with existing habits at scale. Webster’s spellings had to displace an already-automatic reading habit across the entire everyday vocabulary simultaneously. A dozen new pronoun forms are a one-time, front-loaded cost a learner pays once at the start, not a continuous tax on every word they already half-know how to spell. This is a considered bet, not a proven one — it has not been tested empirically the way the LFC tests P3/P4/P7. The honest position is: the coined pronoun paradigms are the part of WoE structurally closest to Webster’s failures, kept deliberately because they are small, systematic, and front-loaded rather than because prior art endorses them.
• Bears on. G4, G12, O5 (the never-coin line, and its one deliberate exception).

C. The empirical base on international intelligibility

The subsets and reforms above are precedent; this section is evidence — research on what actually helps or hurts understanding between non-native speakers. It is where prior art both contradicts and corroborates specific World English rules.

Jenkins’ Lingua Franca Core (LFC) — a real contradiction

• What it was. An empirically-derived set of pronunciation priorities from Jennifer Jenkins, The Phonology of English as an International Language (2000), based on what did and did not cause breakdowns between non-native speakers of English as a Lingua Franca.
• What it found. Core (needed for intelligibility): most consonants, word-initial consonant clusters, vowel length contrasts, and nuclear/contrastive stress placement. Non-core (variation tolerated, treat as accent): the th sounds /θ/ and /ð/, fine vowel quality, lexical word-stress placement, weak forms/schwa, stress-timed rhythm, and attitudinal intonation.
• Where it contradicts WoE. Three places, recorded honestly — and now resolved deliberately: P3 and P4 are kept on reading-aid grounds, P7’s spoken divergence is accepted.
  1. The th split. Jenkins found /θ/ and /ð/ are non-core and safely substitutable (with /t d/, /s z/, /f v/) without harming international intelligibility. Yet P3 keeps and distinguishes them. World English’s justification is on the writing/disambiguation side (the th/dh key marks minimal pairs like breath/breathe) and its scope note already concedes that producing /θ/ stays ordinary learner effort. The LFC says that effort is unnecessary for being understood — and WoE agrees on the spoken axis: a speaker may substitute and still be understood. Resolved: the split is kept because it lives on the reading-aid axis (marking minimal pairs in the key), which the LFC’s spoken finding does not bear on.
  2. Word stress. P4 invests heavily in marking lexical stress on every word; the LFC rates lexical stress non-core (only nuclear/contrastive stress is essential). World English’s marking is still defensible as a reading aid, but it is not, per Jenkins, load-bearing for intelligibility. Resolved: the marking is kept as a reading aid, not claimed as an intelligibility requirement.
  3. Question intonation. The LFC rates grammatical intonation non-core, yet P7 — as redesigned — deliberately load-bears rising intonation for the yes/no question. This one flipped: before the redesign P7 said intonation carries no grammatical load, which the LFC supported; the choice to let questions ride on pitch (rather than coin a particle) turned it into a divergence. World English still treats attitudinal intonation as non-load-bearing (aligned with the LFC); only the question now rides on pitch, and it is mitigated in writing by the leading ?. Resolved: the divergence is accepted — one bounded pitch contrast is worth avoiding a coined question particle.
• Where it agrees with WoE. The LFC rates stress-timed rhythm non-core — direct empirical support for P6 (syllable-timing is fine) — and its treatment of attitudinal intonation as non-core matches P7 for everything except the question. Its emphasis on vowel length over quality aligns with P2 keeping length contrasts (book BUUK vs moon MOON).
• Bears on. P2, P3 (contradiction), P4 (partial contradiction), P6 (support), P7 (partial divergence — question intonation).

Creole & learner-English convergence — strong corroboration

• What it is. English-based creoles and adult learner interlanguage independently simplify English in strikingly similar ways, under the shared pressure of reduced/imperfect input (Bickerton’s Language Bioprogram Hypothesis, 1981/1984, and the World Englishes / ELF literature). ⚠ Bickerton’s strong universalist claim is contested among creolists (substrate and gradualist accounts compete), so this is best read as a robust descriptive tendency, not a settled law.
• The convergent features — several of which WoE independently reinvented:
  • Loss of third-person singular -s → World English M3.
  • Preverbal negation (a single not/no before the verb) → World English G6 (I not like it, Not go!).
  • Invariant / generalized tags (one fixed tag regardless of the main clause) → World English’s invariant G6 tag right?.
  • Regularized / analytic tense-marking (reduced inflection, periphrastic markers, regularized past) → World English M1 (-ed everywhere) and G1 (a leaner tense set).
• Lesson for WoE. These moves are not arbitrary designer choices — they are where English repeatedly lands when simplified by real speakers. That is strong evidence they are learnable and natural, and a good rebuttal to “but that’s not how English works.”
• Bears on. M1, M3, G1, G6.

What the pattern tells World English

D. A design-principle trade-off this evidence surfaces

STE’s rule (“one word, one meaning,” a controlled dictionary) is the closest precedent to design principle 4, “no table of special cases to memorize” — but reading WoE’s own G3 against that precedent surfaces a gap worth recording rather than leaving implicit. G3’s preposition-drop/keep rulings are backed by a per-verb list (31 drop / 4 keep / 2 replace rows in the core lexicon), and G3’s prose says the moves “apply on sight.” STE’s dictionary is exactly this kind of list too — approved/non-approved words a writer looks up — and STE is a success story, not a counterexample; the lesson from STE is that an explicit, looked-up list is a legitimate tool, not a violation, as long as it is stated and small. Where WoE’s practice diverges from the letter of principle 4 is narrower than “there’s a hidden list”: the list is public and versioned (vocabulary Table A), so the acceptance criterion (“no hidden word list”) is met. But “apply on sight” overstates what a learner can actually do without consulting that list — depend on drops its preposition, agree with keeps it, and nothing about the verb’s shape predicts which. That is a real, if narrow, gap between the principle’s letter and the rule’s practice, and it should be named as a trade-off rather than asserted away: WoE accepts an explicit, lookup-based exception list for prepositions (STE’s method) while describing the result in language (“on sight”) that promises more than a lookup system can deliver.

Sources & references

A — Subsets

• Basic English — https://en.wikipedia.org/wiki/Basic_English
• VOA Learning English — https://en.wikipedia.org/wiki/Learning_English_(version_of_English)
• Simplified Technical English (ASD-STE100) — https://en.wikipedia.org/wiki/Simplified_Technical_English ; https://www.asd-ste100.org/about_STE.html (2025 “international standard” claim is vendor-stated)
• Globish — https://en.wikipedia.org/wiki/Globish_(Nerri%C3%A8re)
• Plain English — https://en.wikipedia.org/wiki/Plain_English_Campaign ; https://en.wikipedia.org/wiki/Plain_Writing_Act_of_2010

B — Reforms

• Webster’s reforms — https://www.merriam-webster.com/grammar/noah-websters-spelling-wins-and-fails ; https://en.wikipedia.org/wiki/English-language_spelling_reform
• Simplified Spelling Board / Roosevelt 1906 — https://en.wikipedia.org/wiki/Simplified_Spelling_Board ; https://www.history.com/news/theodore-roosevelt-spelling-controversy (Carnegie dollar figure varies by source)
• Cut Spelling — https://en.wikipedia.org/wiki/Cut_Spelling
• Initial Teaching Alphabet — https://en.wikipedia.org/wiki/Initial_Teaching_Alphabet ; https://www.britannica.com/topic/Initial-Teaching-Alphabet
• Nuclear English — Close, R.A. (1985), “Nuclear English: A Pedagogical Viewpoint,” World Englishes 4(3) — https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1467-971X.1985.tb00373.x ; Quirk, R. (1982), Style and Communication in the English Language (not on Quirk’s Wikipedia page — corroborated via the World Englishes literature)

C — Empirical base

• Jenkins, J. (2000), The Phonology of English as an International Language, OUP — https://en.wikipedia.org/wiki/Lingua_Franca_Core ; https://eprints.soton.ac.uk/65985/
• Creole/interlanguage convergence — https://en.wikipedia.org/wiki/Creole_language ; APiCS (negation typology) https://apics-online.info/parameters/100.chapter.html ; Bickerton, D. (1981) Roots of Language and (1984) The Language Bioprogram Hypothesis (strong universalist claim is contested — treat as a descriptive tendency)

Caveat. Dates and outcomes above were web-verified (2026), but several figures (Carnegie’s SSB funding, STE’s “2025 standard” status) and one theoretical claim (Bickerton’s bioprogram) are flagged as source-dependent or contested. Where prior art contradicts a World English rule — most importantly the LFC on /θ/–/ð/ — the contradiction was recorded honestly and is now resolved deliberately (§C): the diverging rules are kept, on reading-aid grounds for P3/P4 and as an accepted spoken trade-off for P7.