Preface to the Third Edition of the OED

By John Simpson, Chief Editor, OED
The Oxford English Dictionary has been the principal dictionary of record for the English language throughout the lifetime of all current users of the language. The first fascicle or instalment of the Dictionary was published by the Oxford University Press in 1884, and publication continued regularly until the whole text of the First Edition was completed in 1928. After that, two supplements (mainly of nineteenth- and twentieth-century neologisms) were produced, and these were largely incorporated into the Second (unrevised) Edition of 1989. The Dictionary has come to be regarded as authoritative, and in order to maintain its pre-eminence the Delegates of the Oxford University Press decided in 1990 to authorize a comprehensive editorial programme of revision and updating, the preliminary results of which are published here for the first time.

The purpose of the current editorial work on the Dictionary is to produce a completely revised and updated text. A preliminary statement of intent may be found on p. lv of the first volume of the Second Edition (The Future of the OED). Each entry already published is being comprehensively reviewed in the light of new documentary evidence and modern developments in scholarship, and further entries are being added both to fill gaps in the historical record and to record changes in the language today.

The revised and updated text

Each new group of fascicles or instalments of the First Edition of the Dictionary was prefaced by a brief introductory analysis of its contents. The present introduction offers a similar summary of the contents of the first range of newly revised and updated entries (running from the letter M to the word mahurat), as well as some general comments on the policy and scope of the revision.

There was no particular necessity to start the revision work at the beginning of the alphabet. In fact, there were a number of reasons for not doing so, perhaps the most important being that the early sections of the published Dictionary, as with most large-scale historical dictionaries, are characterized by a gradually evolving editorial style and a comparative paucity of documentary evidence. It was therefore decided to start the revision midway through the text, at the letter M, in order to benefit from the editorial stability which the First Edition had achieved by this point.

The first revised and updated instalment consists of 1,045 main entries, containing approximately 2,820 meanings, subordinate words, and phrases. 286 of these main entries have been added since the Second Edition of the Dictionary, though sixty-three first appeared in the Oxford English Dictionary Additions Series (1993, 1997). All entries added from the Additions Series have also been reviewed and wherever necessary revised for online publication.

The text of this range of entries contains about 200,000 words in the Second Edition of the Dictionary; the equivalent revised and updated range contains approximately 400,000 words of text, representing an overall doubling in size.

Back to top

Headwords and the selection of entries

The revised text will include all entries (headwords) and meanings, compounds, phrases, derivatives, etc., included in earlier editions of the Dictionary. It should be noted in this context that the Second Edition of the Dictionary (1989), which was essentially an unrevised conflation of the texts previously published in the First Edition and the Supplements of 1972-86, excluded a small number of brief entries found in the one-volume Supplement to the Oxford English Dictionary of 1933. These are being reinstated in the revised text.

The headword represents the spelling under which a main entry is listed in the Dictionary. Many words display numerous variant spellings over their history, and the headword is either the spelling recognized as the standard contemporary form or (where spelling is currently variable) the most regularly attested modern spelling. For obsolete terms it is normally the form most commonly recorded in the latest period of the word’s history. However, some older terms (as in previous editions of the Dictionary) are listed under unrecorded, modernized spellings, in order to allow them to appear adjacent to later related words. Where there is variation between British and American spelling, the British English form is given first. Standard American English spelling comes immediately afterwards (though none occurs in the first published range).

Reinterpretation of the documentary evidence means that some headwords now appear under different spellings. For example, maccoboy (a variety of snuff) now usually appears as macouba, reflecting a modern respelling after the place name in Martinique where the term originated; maggot-pie (magpie) becomes maggoty-pie, in accordance with later regional evidence which retains a trace of the original form magget the py; and maharanee becomes maharani, reflecting the tendency towards a more ‘scholarly’ transliteration of loanwords from non-roman alphabets. In these cases, the earlier spellings are retained as historical spelling variants and in the illustrative quotations accompanying the entries.

Subsenses in the Second Edition may be conflated with or separated from others to form new integrated or individual subsenses. Many compounds and derivatives previously nested under the parent word are now given independent headword status (e.g. Machiavellianism, maddeningly, magnetic variation, and the verb to mah-jong).

New material (whether historical or modern) is included on the basis of the documentary evidence available to the editorial staff. Further new material, as well as amendments, will be added to published ranges from time to time as the publication schedule allows.

Back to top


The revision of the Oxford English Dictionary‘s etymological component is a substantial undertaking. In the First Edition many entries whose origin was taken to be self-evident (typically native English formations) were not provided with etymologies. In the revised material each entry has a formal etymology. In addition, the names of languages cited in etymologies are now given in full rather than in an abbreviated form and these names have themselves been subjected to an overall review to ensure that the most appropriate modern terminology is used.

References are no longer made to hypothetical reconstructed Indo-European forms. Instead, etymologies refer to recorded cognates formed from the assumed base. The detailed ulterior history of a group of related words is consolidated under the entry for one of these words, with cross-references at the other entries pointing to this discussion.

But the most significant changes relate to the analytical content of the revised etymologies, which for the most part update text which appeared in the First Edition of the Dictionary, and therefore represented the state of scholarly knowledge approximately one hundred years ago. This work has been greatly assisted by the publications (both in book and in machine-readable form) and advice of scholars throughout the world. A notable aspect of the current revision is the availability of Anglo-Norman forms of words which the original editions of the Dictionary regarded as borrowings directly from Central French. This has been made possible largely through the publication of the Anglo-Norman Dictionary, which has provided systematic documentation of this important link between Old and Middle French and (particularly) Middle English. Further as-yet-unpublished medieval and later documentation has kindly been made available to the Oxford English Dictionary by Dr David Howlett, editor of the Dictionary of Medieval Latin from British Sources (currently in progress in Oxford), supplementing material which is available in a much-abridged form in R. E. Latham’s Revised Medieval Latin Word-list. A full list of major dictionaries and related works which have contributed to the revision of the Dictionary will be published in due course. A summary of the current etymological work may be found in Philip N. R. Durkin, “Root and branch: revising the etymological component of the Oxford English Dictionary“, in Transactions of the Philological Society, vol. 97, no. 1, pp. 1-49.

The overall effect of incorporating information on word-forms and meanings from donor languages, together with dates of attestation in those languages wherever possible, is to give a much fuller picture of the process of transmission of individual words into English. These sources also allow the editorial staff to plot contemporaneous borrowings into other languages in Europe and elsewhere, showing that English is part of a network of languages which share in the process of borrowing and semantic development, and that the process of borrowing itself has often been far more complex in particular cases than the Oxford English Dictionary has previously been able to demonstrate.

A typical example of the etymological information which can be adduced from recent work in various languages on the precedents and cognates of an English term is provided by the word magazine. Detailed discussion of the word’s transmission history has not previously been available in the Dictionary. The word magazine derives ultimately from an Arabic term meaning ‘a storehouse’, which appears in a post-classical Latin form magazinus in an Italian document dated 1214. The Italian form magazzino (recorded from 1348) is the source of Middle French magasin (recorded from 1409, and from 1389 in the form maguesin). The English word derives from the French, and is first recorded in 1583, in the sense ‘a place where goods are kept in store’. Many of the later English senses parallel earlier meanings in other European languages, but it is of some interest that the meaning ‘periodical publication’ is an English innovation, not recorded in its French form until later. Needless to say, one of the essential components of a viable etymology for a loanword such as magazine is an established record of cultural contact between speakers of the languages involved, as is here the case with Arabic, Italian, and French. Not surprisingly, the Arabic word also appears in various forms in early Spanish.

Words which begin with the letter M in many ways constitute an interesting cross-section of the English language in terms of its etymological origins. In the first revised range (from M to mahurat) it is perhaps surprising that only two words have been continuously part of the language since Germanic times (mad, though with the loss of an earlier prefix, and madder). Four others are Old English terms; two (mægbot and maegth) have been revived by historians and two (Macedonish and Magnificat) are loanwords. It is notable that there are no certain direct loans from early Scandinavian (‘Old Norse’), although three words are either loans from that source or native formations.

Approximately 420 (40%) of the main entries in the first revised range entered English as borrowings from other languages. A quarter of these were borrowed from classical Latin and from post-classical Latin (using the latter term to embrace all Latin writings from late Antiquity to the present). A further thirty-three are classified as scientific Latin; these are principally taxonomic terms created by modern scientific writers from Latin or Latinate elements. Nineteen words derive from ancient Greek.

There are many borrowings from the post-classical Romance languages. French provides almost seventy (including madam, magic, and magnificent), in the earliest cases dating from the years following the Norman invasion of 1066 and the subsequent establishment of Anglo-Norman rule in the British Isles. Five of these were borrowed from Old French, twenty-nine from Middle French (1325-1600), and a further thirty-five from modern French. In addition, a further ten are borrowed from Anglo-Norman.

Borrowings from elsewhere in Europe are well attested: the Netherlands (five), Italy (fifteen), Spain (sixteen, as well as four from American and Mexican Spanish), Portugal (six), and Germany (twenty-one). Examples include macaroni, machete, madrigal, maestro, mafia, and mahlstick.

The revised entries also provide terms from the native languages of North America (e.g. maccarib, macock, and mahala), the Indian subcontinent (e.g. machan, mahant, mahout, mahua), South Africa (e.g. maanhaar, maas, mabela, mahewu), Australia (e.g. Macquarie, and perhaps mado), New Zealand (mahoe), and elsewhere. In some cases, words which could previously only be ascribed to unspecified indigenous languages or language groups can now be traced more precisely to the language of origin (macaca).

The remaining 60% of main entries arose by various word-formation routes within English. Some 28% arose by suffixation, 4% by the compounding of two independent words, and about 12.5% by prefixation. The remainder arose by other methods (such as blending, back-formation, clipping, etc.) or are of unknown or uncertain origin. (These figures should not be taken as representing the full range or number of word-formations within the revised material, as they do not cover the derivation of subordinate entries and of individual senses which developed subsequent to a word’s arrival in English.)

Back to top


The Dictionary has in the past been criticized for its apparent reliance on literary texts to illustrate the development of the vocabulary of English over the centuries. A closer examination of earlier editions shows that this view has been overstated, though it is not entirely without foundation. The revised text makes use of many non-literary texts which were not available to the original Victorian readers and their immediate successors, particularly social documents such as wills, inventories, account books, diaries, journals, and letters such as the York Civic Records, Gilbert White’s Journals, and the Diaries of Robert Hooke. The inclusion of material from sources such as these allows the editors to provide a fuller picture of the vocabulary of (especially) the Early Modern period. Further reading of similar sources will doubtless result in additional significant discoveries, as will the re-examination of texts already ‘read’ for the Dictionary.

Various factors contribute to the number of quotations that are used to illustrate the history of a particular word or meaning in the Dictionary. In some cases (depending on the length of time a term has been recorded in English) an interval of fifty years between quotations might be appropriate. In others, a longer or shorter time span might be satisfactory. Other significant factors include the relative frequency of the term in a given period, the availability of quotation material, and the need to illustrate numerous spelling variants and grammatical structures.

Approximately 5,800 new illustrative quotations have been added to the first revised range (an increase of 93%), considerably enhancing the documentation of most entries. Almost one in three meanings is now illustrated by one or more earlier examples, some of just a few years, but many of considerably more. Words are sometimes seen now to have arisen at an earlier period of the language than was previously recognized: Old English (such as Magnificat) rather than Middle English, Middle English (such as macaroon, Macedonian, madwoman, and magnificent) rather than Early Modern English, etc.

The Oxford English Dictionary maintains four major reading programmes which make substantial amounts of new data available each year to the editors. The UK and North American Reading Programmes together contribute some two hundred thousand illustrative quotations annually to the Dictionary’s reading-programme database. The texts read for the Dictionary include novels, poetry, diaries, textbooks, newspapers, periodicals, magazines, film and radio scripts, and many other sources, mainly from the nineteenth and twentieth centuries, but in the case of the North American programme, stretching back to the colonial times. The Historical Reading Programme addresses texts from the Early Modern period into the nineteenth century, supplying many antedatings and later examples of words already recorded in the Dictionary, but also identifying words and meanings not yet covered by the Oxford English Dictionary. The Scholarly Reading Programme ensures that scholarly articles relevant to the analysis of the vocabulary of English and published since the First Edition of the Dictionary are monitored for the OED department’s files. In addition, collaboration with the other major historical dictionary projects around the world allows the Dictionary to benefit from material gathered for these dictionaries, and many private contributors make the findings of their own reading and research into the history of the language available to the Dictionary.

The material made available by dictionaries such as the Middle English Dictionary, the Dictionary of American Regional English, and many others has also considerably enhanced the coverage of historical areas and varieties of English in the revised Oxford English Dictionary entries. An extensive file of quotations covering Early Modern English (numbering almost three million quotations) has kindly been made available to the editorial staff of the Dictionary by the University of Michigan. This material was originally collected to form the basis of a proposed Dictionary of Early Modern English (unfortunately the dictionary was not completed: see R. W. Bailey, “Progress toward a Dictionary of Early Modern English 1475-1700”, Proceedings of the Second International Round Table Conference on Historical Lexicography, ed. W. Pijnenburg and F. de Tollenaere, Dordrecht, Holland, 1980, 199-206), and has contributed significantly to the analysis of the language in the Early Modern period represented by the revised text. Examples of first recorded uses deriving from this corpus include machinate (verb), mad-doctor, and magisterious. Again, a full list of major secondary sources consulted by the Dictionary will be published in due course.

Sources only available on CD-ROM or the Internet have been used for the first time in the Dictionary. Online editions of newspapers, for example, are now regularly cited, and large textual databases such as the Chadwyck-Healey Literature Online and the Making of America database of early American texts are monitored for useful material. These and others have considerably enhanced the variety of material cited in the revised entries, and doubtless many other similar sources will become available in the future. However, they are most conveniently used by dictionary editors for subsequent research once a lexical item has been identified for possible inclusion; traditional ‘reading’ is still, in most cases, the most efficient method of making this initial identification, especially when dealing with words having many different meanings, which are very frequently attested in machine-readable collections.

The Dictionary seeks to cite sources which will be accessible to readers in the future. Until a more secure method of archiving references to individual sites on the World Wide Web is possible, the Dictionary will prefer to make use of more established online collections. We are still at an early, and doubtless interim, stage in the accessibility of online sources and this is an area in which policy is likely to change as systems for archiving such resources become more established.

One effect of the extension to the available documentation is to show that many words formerly regarded by the Dictionary as still current are now apparently obsolete (e.g. machopolyp, macrophagocyte, and magnetoscope), and that many which were formerly unlabelled can now be reliably labelled as being rare (e.g. machinization, magnolious, and Mahdism). In all, the revised sample shows 52% more words and meanings marked ‘obsolete’ and 242% more marked ‘rare’ than was the case in the equivalent range of the Second Edition of the Dictionary. It is just as important to monitor how and when terms fade from the language as it is to record their arrival.

The process of adding extensive new documentation (for examples, earlier uses) would sometimes have unbalanced the selection of quotations published to illustrate a particular word or meaning. In these cases, some quotations formerly published in the Dictionary have been silently omitted.

Printed sources are normally dated from their date of publication (for printed texts). Manuscript sources are dated according to the date of the relevant extant manuscript, in the case of Middle English and early Scots material followed by an indication of the actual or presumed date of composition if this is substantially different. Old English sources are dated as either ‘eOE’, ‘OE’, or ‘lOE’ (the first and last indicate ‘early’ and ‘late’ Old English); this is because much of the extant record of Old English appears in late manuscripts and it is not generally possible to guarantee that the particular word under review was not altered or added during the process of manuscript transmission. Dating for Middle English material normally follows that established for the Middle English Dictionary. Although manuscript copies are sometimes available for texts published after the invention of printing, it is normally beyond the scope of this Dictionary to research authorial variation within these manuscripts.

Back to top

Chronology and the historical method

The Oxford English Dictionary is based upon ‘historical principles’, and the meanings of individual words entered in the Dictionary are therefore ordered chronologically, within a semantic framework resembling a family tree. Earlier meanings (or related groups of meanings) of a word are placed before later ones, and it is typically possible to track the semantic development of a word over time throughout an entry.

The First Edition of the Dictionary sometimes imposed a ‘logical’ ordering on the documentary evidence, especially when it was felt that further information, if available, would confirm this interpretation. In the revised material, senses are ordered systematically on the basis of the evidence now available. This has been made possible in large part because a considerably wider body of evidence is now available to the editors. Also, it avoids a tendency to impose formulaic orderings based on proposed semantic hierarchies (e.g. divine; human; animal) which is sometimes apparent in the First Edition of the Dictionary. Needless to say, the mechanics of borrowing and semantic shift do not always work in this formulaic sequence. This principle has not been applied to Old English material, as the amount of surviving literature (much of it in late copies) is too small to allow of reliable chronological interpretation.

An interesting result of this reordering occurs in entries for many Early Modern terms derived from proper names. In the past, the Dictionary tended to present the evidence for the adjective before that for the noun, whether or not the documentary evidence substantiated this. For example, the First Edition orders the adjective Machiavellian before the equivalent noun (in this case, both are presented in the same entry): this is done despite the fact that the noun is recorded earlier. The present revised material reverses the ordering of the parts of speech, showing the noun as the older; this still reflects the nature of the evidence, even though earlier examples have been uncovered for the term in both parts of speech. Similarly, the ordering of the adjective and noun is reversed at Magian, for which an earlier example of the noun has also been found. Madagascarian was previously entered only as an adjective. New evidence shows that the equivalent noun preceded this by a number of years.

It should be noted, however, that main entries which are homographs are listed according to a fixed sequence of parts of speech, as is generally the case in the First Edition of the Dictionary.

Back to top

Varieties of English

From its base in Britain, the English language has expanded over the centuries to become a world language, in which individual varieties share a common core of words but develop their own individual characteristics. When the First Edition of the Dictionary was published, it documented the language of the British Isles in greater detail than the varieties of English which were established or emerging elsewhere. Since that time, a considerable amount of major lexicographical work has been conducted in other areas where English is used, and the current revision is able to benefit from this scholarship. Material from such texts as the Dictionary of American English and the Dictionary of Americanisms, the Dictionary of Canadianisms, the Dictionary of South African English, the Australian National Dictionary, the Dictionary of New Zealand English, and many others, supported by the Dictionary’s own reading programme, has enabled the editors to enhance the coverage of varieties of English worldwide. The English of the British Isles now becomes one (or indeed several) of these varieties, whereas previously standard British English may have been regarded as the dominant form of English.

A large number of words from national varieties of English may be found in the early part of the letter M. Examples of these are listed in the Etymology section above, and many have a long history in their own variety. This trend of inclusion will be continued as editorial work progresses elsewhere in the alphabet. Again, the dating of terms from these varieties illustrates not only the spread of the vocabulary but also the social and cultural changes which necessarily precede or accompany this.

Back to top

Scientific terminology

Extensive work on the history of science by scholars throughout the twentieth century has meant that the Dictionary is now better able to document the development of scientific knowledge and the vocabulary used to describe it. Entries in the Dictionary for scientific words are revised by specialist science editors on the Dictionary’s staff, in conjunction with expert external advisers. New scientific vocabulary is identified from the findings of the Dictionary’s various extensive reading programmes. It should not be surprising that the First Edition of the Dictionary tended to rely for much of its scientific component on a small number of key sources. Additional reading of a far wider range of scientific sources allows the editors to delve more closely into the origins of this vocabulary, and the Dictionary’s own research is supplemented by general academic research into the history of scientific vocabulary (see, for example, R. McConchie, Lexicography and Physicke: Oxford 1997).

The early section of M includes a large scientific component, such as magnetism, magnesium, and magnitude (the apparent brightness of a star). It is interesting to see from the evidence of the vocabulary how the terminology used to describe significant scientific discoveries often seeped into general usage in extended meanings, as poets and others picked up these terms and incorporated them into their own work. Scientific investigation of the properties of magnetism in the late sixteenth and early seventeenth centuries, including the discovery that the Earth itself exerted a magnetic force, had a significant effect on aspects of metaphysical imagery. The first recorded use of the word magnetic occurs in Donne’s Anatomie of the World (1611):

She that had all Magnetique force alone,
To draw, and fasten sundred parts in one.

Detailed bibliographical information is given, wherever possible, for the sources of scientific vocabulary coined outside English. In such instances it is instructive to see when the equivalent term appears to have been first used in English-language sources. Magnolia, for example, was coined in Latin by Plumier after the name of the French botanist Pierre Magnol in 1703, and is first recorded in English after forty-five years in 1748; the name of the mineral magnetite (replacing the earlier Magneteisenstein), however, was coined in German by Haidinger in 1845, and is first found in English six years later in 1851. Parallel developments in other languages tend to demonstrate the expansion of scientific terminology throughout the international scientific community.

Back to top


All definitions have been reviewed during the process of revision and updating. Some have been retained, but many have been rewritten. A number of factors have led to the revision of particular definitions. The principal factor has been the reanalysis of the documentary evidence available for each term, which has sometimes indicated nuances of meaning which were either formerly overlooked (or not present in the language when the entry was previously edited) or which are now seen to be more significant than was previously thought. This applies both to the definition of modern terms and to the definition of historical vocabulary.

As a general rule, definitions should be comprehensible to the general reader. This is problematic in the case of some technical vocabulary. However, wherever possible a technical definition should include some pointers which allow the general reader either to understand the type of word being described, or to appreciate the context in which it is used. At times, the supporting illustrative quotations throw further light on these aspects.

Any supplementary information relevant to the definition (perhaps of a historical or explanatory nature) is given in small type (or parentheses) after the definition.

A consistent approach has been adopted to the separation of definitions from the associated text sometimes used to introduce or extend these. In general, ancillary text is closed off by punctuation (usually colons) or by parentheses.

Changes in formal terminology have also led to systematic revisions: some places are nowadays known by different names than they were when the Dictionary was first edited (see, for example, Macedonian); the extension of the metric system and alterations in the terminology of national currencies have promoted further systematic editorial changes (the Chinese mace as a denomination from 1890 was not noted in earlier editions of the Dictionary); taxonomic terminology changes over time and these changes need to be reflected in revised entries. Other types of alteration will be apparent throughout the revised material. Sometimes, a simple substitution of terminology is all that is necessary; at others, the former name is significant to the development of the word under review, and it is necessary to introduce the modern term while retaining a reference to the older one.

Back to top

Variant spellings

Many words are recorded in a number of different spellings in the documentary evidence available to the editors. As comprehensive a list as it has been practical to compile is included in entries which demonstrate such variation. Individual forms are preceded by a note of the century or centuries in which they are recorded, and when necessary by a label indicating the variety of English to which they are restricted. Centuries of use are presented according to the first two digits of the century in question (e.g. ‘17’ for the eighteenth century, as opposed to ‘8’ in the First Edition of the Dictionary). Absence of a hyphen in a range of dates implies discontinuity, e.g.’15 17-18′ (i.e. attested in the 16th century and in the 18th and 19th centuries, but not in the 17th century).

Century dates are not used for forms from Old English (pre-1150) and Middle English (1150-1499) or (in general) from Older Scots (pre-1700). Older Scots forms are listed as ‘pre-17’ (except in occasional, typically isolated and late, examples where the date is secure). Old and Middle English forms are listed as ‘OE’ or ‘ME’, preceded by ‘e’ or ‘l’ for ‘early’ or ‘late’ when appropriate.

The default designation for all Middle English forms is ‘ME’. Within the Middle English period, if a word becomes obsolete before c1325, its forms are labelled ‘eME’, and, conversely, if a word is not attested before 1400, its forms are labelled ‘lME’. In addition, any form (of any word) that is deemed characteristically ‘early’ or ‘late’ may be labelled, respectively ‘eME’ or ‘lME’.

The default designation for all Old English forms is ‘OE’. Within the Old English period, if a word is not attested before 1100, its forms are labelled ‘lOE’. In addition, any form (of any word) that is deemed characteristically ‘early’ (pre-955) or ‘late’ (1100-1149) may be labelled, respectively ‘eOE’ or ‘lOE’.

In the First Edition a number of truncated forms are given; these have been expanded to the full form wherever possible in the revised material. The typical expansion of this aspect of the revised entries may be seen by comparison with the data provided in the First Edition:

Forms: 3-6 makerel(l, 5 makerelle, makyrelle, 4-7 macrel(l, makrel(l, 5 macrelle, 6 macquerell, 7 maquerel, 7-8 macril(l, maycril, 6-8 mackrell, 7-9 mackrel, 4-9 mackerell, 7-9 mackarel, 8 mackarell, 7- mackerel.

(First Edition: mackerel)

Forms: ME macrelle, makerelle, makyrelle, ME-15 makarell, makerell, ME-16 macrel, makrel, makrell, ME-17 macrell, ME-18 mackerell, makerel, 15 macquerel, macquerell, 15-17 mackarell, mackrell, 16 maquerel, 16-17 macril, macrill, maycril, 16-18 mackaral, mackarel, mackrel, 16- mackerel, 17 maccarel, mackril, 18- mackeral; Sc. pre-17 mackreel, pre-17 macrell, pre-17 makarell, pre-17 makcaral, pre-17 makral, pre-17 makrall, pre-17 makreill, pre-17 makrel, pre-17 makrell, pre-17 17 mackrel, 17- mackerel, 18- macrel.

(Revised entry: mackerel)

Back to top


The modern pronunciations and stress patterns given in the original edition of the Dictionary were represented according to a system of notation devised by the first editor, Sir James Murray, and reflected educated southern English pronunciation at the end of the nineteenth century. This notation system was translated in the Second Edition, largely symbol by symbol, into an equivalent transcription using the International Phonetic Alphabet (which had originally been devised slightly too late to be available to the Dictionary when it began publication in 1884).

Each pronunciation in the revised text is given in the International Phonetic Alphabet (IPA), according to a revised model of Received Pronunciation devised by Dr Clive Upton of the University of Leeds, and the scope of this information has been extended to include a ‘standard’ U.S. pronunciation based on a model devised by Professor William Kretzschmar of the University of Georgia. Words from other varieties of English have been given pronunciations based on (but not identical in all details with) the models used by the principal historical dictionaries of World Englishes (such as the trisyllabic New Zealand term mahoe, originally from Maori, and the disyllabic Caribbean mahoe). The Dictionary does not aim to cover dialectal variation in pronunciation within each variety.

When the modern British English pronunciation of a word is a direct descendant of that documented in the First or Second Edition of the Dictionary, the differences are not systematically noted. However, when the current British English pronunciation differs from the earlier form, the pronunciation given in the earlier edition is recorded in an editorial note. Detailed information on significant historical irregularities and peculiarities in the pronunciation of individual words is given in etymological notes. This is extended where appropriate to include discussion of historical pronunciation (e.g. as evidenced by variant spellings), as at machete and Magdalen.

Back to top

Bibliography and textual accuracy

The Dictionary’s definitions are informed by the evidence of the language collected by thousands of ‘readers’ from the mid nineteenth century to the present day. Much additional information has also been provided by scholars and others through their publications and correspondence with the Dictionary’s editorial staff. Nowadays, documentary information is also available from computerized sources: concordances, large full-text databases, etc.

Many of the quotations published in earlier editions of the Dictionary will also be found in the revised material. Also, many new quotations have been added. It is important that these represent the texts as originally published, and in order to achieve this, many quotations are being reverified in more reliable editions than those from which they were originally collected. A typical example might be a sixteenth-century text ‘read’ by the Dictionary in a modernized and amended version, perhaps edited by a Victorian scholar from a late edition.

Over the course of the revision, many thousands of texts will be re-examined in this way, and the results fed text by text on to the online database. Reverification takes place principally by printing out all of the quotations from a particular source (typically illustrated throughout the Dictionary) and comparing these against the appropriate edition, making amendments when necessary. At present, 2700 texts have been re-examined in this way. The process of reverification will continue through the period of revision, and doubtless beyond, and will affect entries already published online in their revised and updated form. This is one reason why revised entries are marked as ‘draft’ versions and not as published in a final form. A typical example might be the following, which shows the version of a quotation as represented in the First Edition of the Dictionary, and the text as amended after reverification. Such changes will occur throughout the revised material.

1509 BARCLAY Shyp of Folys (1570) 206 For his strength and magnanimitie..One founde on grounde like to him can not be.

(First Edition: magnanimity)

1509 A. BARCLAY tr. S. Brant Shyp of Folys f. ccxviiv, For his strength and magnanymyte..One founde on grounde lyke to hym can nat be.

(Revised entry: magnanimity)

The conversion of the text from the edition of 1570 to the first edition of 1509 shows several differences in spelling, both of the headword itself (from magnanimitie to magnanymyte) and of words in the surrounding text. The revised bibliographical information associated with the quotation shows both Barclay’s initial and the author of the original work of which Barclay’s text was a translation.

Similar improvements across the database occur through the process of rechecking the bibliographical information associated with each quotation. Advances in scholarship mean that bibliographical information available to the original editors of the Dictionary is sometimes no longer considered accurate. This may concern the dating of texts, the authorship ascribed to them, the style by which they are cited, and many other features. As with so much in historical lexicography, this bibliographical enhancement is a potentially never-ending activity, and the reader will become aware that changes in the bibliographical data are being made gradually to the online database over the course of the revision project. In this and other ways the revised Dictionary will remain a ‘moving document’, susceptible of improvement and alteration as scholarship advances and as the language continues to change.

Back to top


There are a number of myths about the Oxford English Dictionary, one of the most prevalent of which is that it includes every word, and every meaning of every word, which has ever formed part of the English language. Such an objective could never be fully achieved. The present revision gives the editors the opportunity to add many terms which have been overlooked in the past, but it should be understood that fully comprehensive coverage of all elements of the language is a chimera. That said, the content of the Dictionary is certainly comprehensive within reasonable bounds.

It is also often claimed that a ‘word’ is not a ‘word’ (or is not ‘English’) unless it is in ‘the dictionary’. This may be acceptable logic for the purposes of word games, but not outside those limits. Proponents of this view expect dictionaries to include ‘proper’ English, whereas dictionaries in fact include many slang, informal, technical, and other words which such people might not consider to be ‘proper’, typically labelled according to the register of language to which they belong. It may be added here that the question ‘How many words are there in the English language?’ cannot be answered by recourse to a dictionary.

Another myth about the Dictionary, and about dictionaries in general, is that they provide a comprehensive analysis of each word treated. Again, this cannot be the case in a finite text. But more important, philosophically, is that any dictionary attempts to provide information in a manner which is accessible to the reader. In order to do this, it is customary to subdivide polysemous words by their meanings and by the grammatical and syntactic forms in which they are found. However, any extensive examination of the documentary evidence for a language soon uncovers examples of usage which straddle two or more of the stated meanings of a word, often idiosyncratically and in ways which it is not practical for the dictionary to illustrate. The reader should be aware of this incongruity, and should regard the Dictionary as a convenient guide to the history and meaning of the words of the English language, rather than as a comprehensive and exhaustive listing of every possible nuance.

Readers who have had access to the CD-ROM version of the Dictionary in the past have expressed considerable interest in the number and type of quotations from individual sources. The fact, for example, that Shakespeare was cited over 30,000 times in the Second Edition of the Dictionary has been adduced as an index, amongst other things, of Shakespeare’s lexical creativity. It is probably true to say that too much weight has been given to such statistical theories. First citations from an author are of interest in that they represent the earliest recorded evidence for a particular term. But in many cases this does not mean that a word or meaning was first used in the source cited, but rather that no earlier evidence has yet come to light. Printed evidence will typically not exemplify the first-ever use of a term (except for some self-conscious coinages), but only the first that has been found in the historical record. Later quotations for a particular word or meaning are selected from the sum of those available to the editors, but often reveal more about the reading programme for the Dictionary and the selection criteria of the editors than about the lexical significance of the source.

The points above are not intended as a disclaimer, but as a corrective to those who expect more of a dictionary than it is bound to deliver. It is hoped that the contents of the revised and updated text of the Dictionary will give a clearer picture of the history of the English language through the development and meaning of its vocabulary than has been possible in previous editions. Advances in scholarship over the last one hundred years provide ample material for revising the text, and the complementary work of the editorial staff and their associates has also uncovered a mass of documentary and other evidence which allows us now to offer what may be seen as an appropriately detailed and comprehensive analysis of the English language as it has developed over the past one thousand years and more.

John Simpson, Chief Editor, Oxford English Dictionary
Oxford, March 2000