ictionaries have been a feature of our lives for so long they are almost a fixture on our bookshelves; it’s hard to imagine a time when they did not exist. In fact, the need for dictionaries has been with us almost as long as civilisation itself. In 2000bc, the conquerors of ancient Sumer (roughly present-day Iraq) produced lists of Sumerian words translated into their own language, on tablets of clay, to help with communication. More than 3,000 years later, the first dictionaries to appear in England were also bilingual, and also intended as practical aids to understanding.

The first English work to be called a dictionary, the Dictionary of Sir Thomas Eliot Knight, was published in 1538. It translated Latin words into English for the benefit of the rising middle class, who did not have a classical education and needed help with Latin and Greek. The first wholly English dictionary, one that would be vaguely recognisable to us, did not appear until 1604. Robert Cawdrey’s Table Alphabeticall consisted of 2,500 “hard words” with their everyday English equivalents. “Hard words”, otherwise known as “inkhorn terms”, were obscure words of Greek and Latin origin that, as Cawdrey put it, “Ladies, Gentlemen or other Unskillful Persons... shall heare or read in Scriptures, Sermons or elsewhere”.

Some of these opaque words were the invention of dictionary-makers themselves, the result of translating a Greek or Latin word by simply adding an English ending to it. It says much about the conservative nature of dictionaries that these inkhorn terms persisted for more than 100 years despite being rarely if ever used. Some typical examples are stultiloquence (foolish talk), supervacaneousness (needlessness), and nullibiety (the state of being nowhere).

By the 17th century there was a growing demand for a book that standardised the English language as it was actually used. Imagine a world where Shakespeare and his contemporaries were writing with no comprehensive authority on English to refer to, one reason, perhaps, for their lack of inhibition about inventing new words. It was not until the 18th century that dictionary-makers began to include everyday words and weed out the weighty-sounding inkhorn terms. Then, as now, opinion-formers worried about the language degenerating, and some hankered for an English equivalent of the Academie francaise across the Channel. It was against this background that Samuel Johnson began writing his famous dictionary, the first truly standard work, and the first to be evidence-based.

Johnson’s Dictionary of the English Language took a mere nine years to write and was published in 1755, 250 years ago. It was a truly monumental effort. Johnson worked with a team of six assistants, but they were mainly copyists; the labour of writing the definitions was his own, and astonishing. He was innovative in that he started out by analysing the best writers in the English language: he collected quotations from them first, and only then wrote definitions based on the material he had gathered. For the first time, a dictionary-maker was recording how the written language was actually used, in a way that was not arbitrary or subjective.

And although Johnson began by wanting to “fix” the language in midstream, by the end of his project he was less naive. In his preface he acknowledges the forces behind language change, debunking the lexicographer who hopes to “embalm his language, and secure it from corruption and decay”.

Unwittingly, Johnson set a pattern for lexicographers: a century later the Oxford English Dictionary followed the same principles, only this time with dozens of compilers, and hundreds of volunteers collecting words. Due to Johnson’s influence the OED was biased, some would say excessively, towards words that were culled from carefully selected literary sources. Nowadays lexicographers use electronic corpora - huge databases containing millions of words of text - to research English usage. These- databases include a wide-ranging and balanced selection of different types of English, not just literature but also journalism, broadcast speech and even transcripts of informal conversation.

Computers have also taken over laborious tasks, such as alphabetisation, and provide sophisticated statistical evidence on the most frequent words and meanings in the language. But dictionary editors today still essentially follow Johnson’s method, starting from the evidence and working towards the definition.

There is, however, one significant difference between modern dictionaries and Johnson’s. He was sole author and creator, and he put his stamp on his work in many individual ways. Today’s dictionary writers strive to be neutral and even invisible - describing, not prescribing, what is good or bad in the language. Johnson is largely objective but allows himself the odd personal opinion. He defines shabby as “a low word that has crept into conversation and low writing; but ought not to be admitted into the language”. Of gaol he says: “It is always pronounced and too often written jail or goal.” He indulges in the occasional sideswipe - famously at the entry for oats (the food of Scots and horses), and pension: “In England it is generally understood to mean pay given to a state hireling for treason to his country.”

He is also refreshingly open about the limits of his knowledge. Trollop is “a low word, I know not whence derived”. The definition of trolmydames says simply, “Of this word I know not the meaning.” In his preface he writes, with disarming honesty: “Some words there are which I cannot explain, because I do not understand them; these might have been omitted very often with little inconvenience, but I would not so far indulge my vanity as to decline this confession”’.

Johnson’s transparency and his admission of less-than-perfect knowledge are (perhaps sadly) no longer part of dictionary tradition. Yet despite his modesty and its acknowledged gaps, the Dictionary became the standard reference work, until late Victorian times. Since Johnson, dictionaries have been viewed as authoritative, an impression reinforced by the anonymity of later works. Compiled by large teams, with specialist editors focusing on areas such as science or grammar, they are collective and impersonal efforts. Lexicographers today are trained to avoid letting personal opinions intrude on their definitions. Nonetheless, dictionaries do tend to have their own distinctive character, usually the result of style policies hammered out by the editorial team. Dictionaries, like everything else, have their fashions and their idiosyncrasies.

Johnson’s Dictionary gives us a window onto the mid-18th century - its health, transport, plants, science, education, philosophy, religion, business, food, art, games and social ranks. Most of all, it shows the language in the process of changing. We see words that are still foreign to Johnson, which he puts in italics: scalpel, gusto, opera, lingo, ruse (“a French word neither elegant nor necessary”). We see words which he tells us are dying out, such as pinguid, meaning “fat”. He shows us new meanings that are coming in, such as the modern sense of magazine: “Of late this word has signified a miscellaneous pamphlet.” He includes words we might not expect to find in a book written at this time: ginseng, alimony, grenade, electricity, atom, suburb, mango (“a fruit of the isle of Java, brought to Europe pickled”). Some familiar words have unfamiliar meanings: a go-cart is “a machine in which children are inclosed to teach them to walk, and which they push forward without danger of falling”. Factory means “a house or district inhabited by traders in a distant country”; lunch is “as much food as one’s hand can hold”.

Johnson fretted about the intrusion of Italian and French words into English: today, some feel equally strongly about American English. We may smile at Johnson tut-tutting over “shabby”, but every age has its own examples of language misuse. Looking at a dictionary from a previous age is a salutary reminder that yesterday’s bad usage often becomes today’s standard English - or the reverse. For example, the growing use of “disinterested” to mean “uninterested” loses, in the opinion of some, the valuable meaning of “not self-interested”. Yet originally this meaning of “uninterested” was the correct, prime use of the word. Many still object to split infinitives, although there is no logical basis for this in grammar.

And the use of hopefully at the beginning of a sentence raises hackles, although it is performing exactly the same function as adverbs such as luckily, mercifully and thankfully, which do not attract criticism. As a result, dictionary-makers nowadays tend not to sit in judgment but simply reflect attitudes to usage, issuing carefully neutral warnings such as “Some people object to the use of this word”. Like Johnson, today’s lexicographers accept that language change is inevitable: no Academy could ever stem the tide.

There is one final way in which Johnson blazed a trail for the modern dictionary. He was aware that foreigners would be using his dictionary and deliberately included information that would be helpful to them - for example, two or three-word verb combinations such as “break off” and “put up with”, which mean more than the sum of their parts. Here Johnson was anticipating a development of the mainstream dictionary, namely the dictionary designed for foreign learners of English.

This British publishing phenomenon began in the 1940s, and as the number of learners of English around the globe has mushroomed - about one billion at recent estimates - so these learners’ dictionaries have prospered, outselling their native-speaker stablemates many times over. What makes them different is the additional information they give on grammar and how words combine together. For example, a learner of English has to be told that you commit a crime, not make or do it, whereas a native speaker of English knows this intuitively. Learners’ dictionaries have also been at the forefront of lexicographic innovation, for example in using corpora to provide evidence, or in using a limited vocabulary to make definitions easier to understand. With an ever-expanding market of people wanting to speak English, they are the real growth area in dictionary publishing.

Like Johnson, we are living through a time of great technological and scientific change, which in turn has had dramatic effects on our language.

New words pour into English faster than ever before, propelled by the international media. Dictionaries are now revised far more frequently than before, partly because of improvements in typesetting and printing, but partly because of the different expectations of the electronic age: up-to-dateness is everything.

Most dictionaries are now available on CD-Rom or online, and these electronic editions challenge the conventions of the print format. Powerful search software makes alphabetical order an anachronism, and space-saving features such as abbreviations become redundant, while cross-references can be accessed by the click of a mouse.

Increasingly the print dictionary is being viewed as an endangered species - and in the age of Google and the web, even commercial electronic dictionaries may be under threat. The traditional view of the dictionary as being like the Bible, something fixed, authoritative, and impersonal, is changing; the dictionary of the future will be more dynamic, interactive and open-ended. The ambitious and idealistic Wiktionary project (see panel), which invites us all to become lexicographers, may be a prototype of things to come.

Faye Carney is editor of the new edition of the Concise Bloomsbury Dictionary, published by AC Black (pound;19.99)