Redefining Definition
By ERIN McKEAN
Published: December 17, 2009
If anything is guaranteed to annoy a lexicographer, it is the journalistic habit of starting a story with a dictionary definition. “According to Webster’s,” begins a piece, blithely, and the lexicographer shudders, because she knows that a dictionary is about to be invoked as an incontrovertible authority. Although we may profess to believe, as the linguist Dwight Bolinger once put it, that dictionaries “do not exist to define but to help people grasp meanings,” we don’t often act on that belief. Typically we treat a definition as the final arbiter of meaning, a scientific pronouncement of a word’s essence.
But the traditional dictionary definition, although it bears all the trappings of authority, is in fact a highly stylized, overly compressed and often tentative stab at capturing the consensus on what a particular word “means.” A good dictionary derives its reputation from careful analysis of examples of words in use, in the form of sentences, also called citations. The lexicographer looks at as many citations for each word as she can find (or, more likely, can review in the time allotted) and then creates what is, in effect, a dense abstract, collapsing into a few general statements all the ways in which the word behaves. A definition is as convention-bound as a sonnet and usually more compact. Writing one is considered, at least by anyone who has ever tried it, something of an art.
Despite all the thought and hard work that go into them, definitions, surprisingly, turn out to be ill suited for many of the tasks they have been set to — including their ostensible purpose of telling you the meaning of a word. Overly abstract definitions are often helpful only if you come to them already primed by context. It’s difficult to read a definition like “(esp. of a change or distinction) so delicate or precise as to be difficult to analyze or describe,” and have subtle immediately spring to mind; or to come across “reduce the force, effect or value of” and think of attenuate.
Definitions are especially unhelpful to children. There’s an oft-cited 1987 study in which fifth graders were given dictionary definitions and asked to write their own sentences using the words defined. The results were discouraging. One child, given the word erode, wrote, “Our family erodes a lot,” because the definition given was “eat out, eat away.”
Neither are definitions complete pictures of all the possible meanings of a word. One study found that in a set of arbitrarily chosen passages from modern fiction, an average of 13 percent of the nouns, verbs and adjectives were used in senses not found in a large desk dictionary. And of course there are some words that simply elude definition, a problem even Samuel Johnson faced. In the preface to his groundbreaking Dictionary of the English Language, he wrote, “Ideas of the same race, though not exactly alike, are sometimes so little different that no words can express the dissimilitude, though the mind easily perceives it when they are exhibited together.” We all have had Johnson’s experience of “easily perceiving” differences between words that we cannot as easily describe — quick: what’s the difference between louche and raffish? Most people, when asked what a word means, resort to using it in a sentence, because that’s the way we learn words best: by encountering them in their natural context.
Given these shortcomings of definitions, and the advantages of examples, why do we still cling to definitions? The short answer, for hundreds of years, has been a practical one: space — specifically the lack thereof. Print dictionaries have never had sufficient page-room to show enough real, live, useful examples to create an optimal and natural word-learning experience. Even the expert lexicographers at the Oxford English Dictionary, which famously includes “illustrative quotations” alongside its definitions, still put the definition and its needs first, making new words wait their turn to make it through the definition bottleneck.
The near-infinite space of the Web gives us a chance to change all this. Imagine if lexicographers were to create online resources that give, in addition to definitions, many living examples of word use, drawn not just from literature and newspapers but from real-time sources of language like Web sites, blogs and social networks. We could build people’s confidence in their ability to understand and use words naturally, from the variety of contexts in which words occur. Indeed, this is what my colleagues and I are trying to accomplish at the online dictionary Wordnik.com: we’re using text-mining techniques and the unlimited space of the Internet to show as many real examples of word use as we can, as fast as we can.
This approach is especially useful for grasping new words and uses: if you look up tweet on a site like mine, for example, you understand that the word is used to refer to messages sent via Twitter; there’s no waiting for an editor to write you a definition; plus there are examples of tweets right on the page. Online, you can also look up just the form of a word you’re interested in — say, sniped instead of snipe — and find precise examples. A word is so much more than its meaning: it’s also who uses it, when it was used, what words appear alongside it and what kinds of texts it appears in.
Without privileging definitions, dictionary-making would involve more curation and less abridgment, less false precision and more organic understanding. If we stop pretending definitions are science, we can enjoy them as a kind of literature — think of them as extremely nerdy poems — without burdening them with tasks for which they are unsuited.
Erin McKean is the chief executive and founder of the online dictionary Wordnik.com. She was previously the editor in chief of American Dictionaries for Oxford University Press.
By ERIN McKEAN
Published: December 17, 2009
If anything is guaranteed to annoy a lexicographer, it is the journalistic habit of starting a story with a dictionary definition. “According to Webster’s,” begins a piece, blithely, and the lexicographer shudders, because she knows that a dictionary is about to be invoked as an incontrovertible authority. Although we may profess to believe, as the linguist Dwight Bolinger once put it, that dictionaries “do not exist to define but to help people grasp meanings,” we don’t often act on that belief. Typically we treat a definition as the final arbiter of meaning, a scientific pronouncement of a word’s essence.
But the traditional dictionary definition, although it bears all the trappings of authority, is in fact a highly stylized, overly compressed and often tentative stab at capturing the consensus on what a particular word “means.” A good dictionary derives its reputation from careful analysis of examples of words in use, in the form of sentences, also called citations. The lexicographer looks at as many citations for each word as she can find (or, more likely, can review in the time allotted) and then creates what is, in effect, a dense abstract, collapsing into a few general statements all the ways in which the word behaves. A definition is as convention-bound as a sonnet and usually more compact. Writing one is considered, at least by anyone who has ever tried it, something of an art.
Despite all the thought and hard work that go into them, definitions, surprisingly, turn out to be ill suited for many of the tasks they have been set to — including their ostensible purpose of telling you the meaning of a word. Overly abstract definitions are often helpful only if you come to them already primed by context. It’s difficult to read a definition like “(esp. of a change or distinction) so delicate or precise as to be difficult to analyze or describe,” and have subtle immediately spring to mind; or to come across “reduce the force, effect or value of” and think of attenuate.
Definitions are especially unhelpful to children. There’s an oft-cited 1987 study in which fifth graders were given dictionary definitions and asked to write their own sentences using the words defined. The results were discouraging. One child, given the word erode, wrote, “Our family erodes a lot,” because the definition given was “eat out, eat away.”
Neither are definitions complete pictures of all the possible meanings of a word. One study found that in a set of arbitrarily chosen passages from modern fiction, an average of 13 percent of the nouns, verbs and adjectives were used in senses not found in a large desk dictionary. And of course there are some words that simply elude definition, a problem even Samuel Johnson faced. In the preface to his groundbreaking Dictionary of the English Language, he wrote, “Ideas of the same race, though not exactly alike, are sometimes so little different that no words can express the dissimilitude, though the mind easily perceives it when they are exhibited together.” We all have had Johnson’s experience of “easily perceiving” differences between words that we cannot as easily describe — quick: what’s the difference between louche and raffish? Most people, when asked what a word means, resort to using it in a sentence, because that’s the way we learn words best: by encountering them in their natural context.
Given these shortcomings of definitions, and the advantages of examples, why do we still cling to definitions? The short answer, for hundreds of years, has been a practical one: space — specifically the lack thereof. Print dictionaries have never had sufficient page-room to show enough real, live, useful examples to create an optimal and natural word-learning experience. Even the expert lexicographers at the Oxford English Dictionary, which famously includes “illustrative quotations” alongside its definitions, still put the definition and its needs first, making new words wait their turn to make it through the definition bottleneck.
The near-infinite space of the Web gives us a chance to change all this. Imagine if lexicographers were to create online resources that give, in addition to definitions, many living examples of word use, drawn not just from literature and newspapers but from real-time sources of language like Web sites, blogs and social networks. We could build people’s confidence in their ability to understand and use words naturally, from the variety of contexts in which words occur. Indeed, this is what my colleagues and I are trying to accomplish at the online dictionary Wordnik.com: we’re using text-mining techniques and the unlimited space of the Internet to show as many real examples of word use as we can, as fast as we can.
This approach is especially useful for grasping new words and uses: if you look up tweet on a site like mine, for example, you understand that the word is used to refer to messages sent via Twitter; there’s no waiting for an editor to write you a definition; plus there are examples of tweets right on the page. Online, you can also look up just the form of a word you’re interested in — say, sniped instead of snipe — and find precise examples. A word is so much more than its meaning: it’s also who uses it, when it was used, what words appear alongside it and what kinds of texts it appears in.
Without privileging definitions, dictionary-making would involve more curation and less abridgment, less false precision and more organic understanding. If we stop pretending definitions are science, we can enjoy them as a kind of literature — think of them as extremely nerdy poems — without burdening them with tasks for which they are unsuited.
Erin McKean is the chief executive and founder of the online dictionary Wordnik.com. She was previously the editor in chief of American Dictionaries for Oxford University Press.
I would probably have said that the difference between "louche" and "raffish" is that I have, at least once, heard of "raffish".
ReplyDeleteGreat article, and wonderful website (Wordnik). Thanks for this, Stan.
ReplyDelete