Words And Their Meanings And The Role Of A Dictionary

Words And Their Meanings And The Role Of A Dictionary

One of the crucial features of the intelligent reading machine for the blind and sight-impaired that I was proposing was the ability to read text for me (not just to me) and tell me what the text was about. I was confronted with the hard question of how to teach a machine how to understand what it reads. The approach that I wanted to take was to model the way that humans understand words and their meanings. In this pursuit, I became interested in how words were defined in dictionaries.

There are two general schools of thought regarding the role of a dictionary:

1. A dictionary tells us what words mean and how to use them correctly. When used in this way, a dictionary is an authority of language use. It is a guide to help us to communicate effectively using a common language.

Samuel Johnson produced (by himself) “A Dictionary Of The English Language” that he published in 1755, It took him 8 years to complete and marked an important step in English lexicography. He attempted to address what he considered to be deficiencies in dictionaries of English at the time. The preface of his dictionary describes his methodology for collecting and defining words and for formalizing rules of grammar and word usage, stating that “the duty of the lexicographer is to correct and prescribe” and to give structure to the conventions of the English language. He announced his project in “The Plan Of An English Dictionary” in 1747 where he explained how etymology can be used to separate long-established words in the language from new words created from common practice rather than from derivation.

Johnson’s strict view was likely due in part to the limitations of his resources. For instance, he explained that certain verbs, such as ‘do’, ‘go’, and ‘make’, that are used frequently and have meanings that are “loose and general” or whose usage is “vague and indeterminate” would require an enormous amount of effort to document properly. He acknowledged later that “terms of art” (technical terms) were omitted due to lack of time; he elaborated, “I could not visit caverns to learn the miner’s language, nor take a voyage to perfect my skill in the dialect of navigation”. The resulting dictionary was a guide to the correct use of English words.

Others have seen value in central sets of rules to describe conventions for language use. For example, Henry Fowler, in 1926, published “A Dictionary Of Modern English Usage” as a style guide for writing, as well as pronunciation and word use. After all, if the primary purpose of language is to communicate, then shouldn’t the role of a dictionary be to help us to communicate better? But what happens when people use words in ways, and with meanings, not defined in the dictionary?

Language change happens from generation to generation in every living language. Anyone who has been speaking or reading for a few decades has surely noticed this from personal experience. Each time there is an advance in technology, or science, or tools, the new innovation comes with new terminology that often uses words already in existence. For example, Alan Turing, in 1936, introduced the mathematics behind what would become the digital computer, now known simply as the computer. The word “computer” was not created for this purpose. In fact, Noah Webster’s “An American Dictionary Of The English Language” from 1828 already had a definition for the word “computer”, namely “One who computes; a reckoner; a calculator”.

Variations in language use are more extreme when a new method of communicating is introduced. Those who remember when the world began corresponding with emails instead of sending letters in the post may have noticed their own writing style change. The effects of text messaging were even more abrupt.

These points introduce a second view of the role of a dictionary.

2. A dictionary tells us how words ARE used in practice. One word may have many different definitions (or senses). Each sense of a word represents one of the ways in which that word can be used to convey meaning.

Richard Chenevix Trench, Archbishop of Dublin, gave a lecture to the Philological Society in London on November 5, 1857, that laid out criticisms of existing dictionaries and presented a plan for a new approach to creating an English dictionary. This lecture was published under the title “On Some Deficiencies In Our English Dictionaries” and prompted the Philological Society to begin the project that is now the Oxford English Dictionary.

Trench described a dictionary as “an inventory of the language”. The role of the dictionary maker, in this view, is not to “select the good words of a language”. He contrasted his view of the purpose of a dictionary with existing dictionaries which were designed to be “a standard of the language”. Trench’s vision for a dictionary would rely on dozens of volunteers to document words and their meanings in the contexts in which they were used. This is the opposite of the approach used by Samuel Johnson, who collected and defined all of his words himself.

The Oxford English Dictionary (OED) was created largely using Trench’s basic idea, and invitations were published for any reader to submit words and citations of where and how the words were used in print. This gave the OED access to a wealth of historical information about the use of nearly 500,000 words. An important drawback to this approach was that it took more than 70 years to compile the first complete version of the OED, released in 1928.

Noah Webster announced his plan in 1789 to create an American dictionary of English. He described his project in that year with the publication of “Dissertations On The English Language” where he argued for the need of an American dictionary that was distinct from the British dictionaries. He urged that, “As an independent nation, our honor requires us to have a system of our own, in language as well as government”.

Webster defended the importance of “common practice” in describing the meanings and contexts of a word. In contrast, he said that few dictionary grammar rules written to prescribe the correct use of words “can be vindicated on any better principle than some Latin rule, or his [the grammar writer’s] own private opinion”. Although Webster believed that a dictionary should document the way words are used in practice, he limited his scope by excluding words and meanings that were only used locally. He said that the chief job of the compiler of a grammar is “to separate local or partial practice from the general custom of speaking”. That is, the compiler must identify which patterns of word use correspond to large communities and which may only be used by a few. He also took this point of view regarding the pronunciations of words appearing in the dictionary, saying that “the common consent of a nation is sufficient to stamp it with propriety”. Webster completed his dictionary in 1828 and published it as “An American Dictionary Of The English Language”.

The Merriam-Webster dictionary today, along with other English dictionaries, continue to combine the two schools of thought. New words are added to the dictionary every year, along with new senses of existing words. This gives us a living guide to the way words are being used in our present day. Although the benefits of this practice may be evident, some of the additions may be controversial, particularly when new patterns of use appear to make a useful word less meaningful. An example of such an addition is the word “literally”. Merriam-Webster defines a second sense of the word “literally” as “in effect : virtually”. In this sense, the word “literally” is used to emphasize a point being made with a metaphor, as in the sentence “I am literally dying of boredom”.

Words in Skimcast

As I was developing the theory behind what is now Skimcast, I saw two possible options for enabling a machine to acquire knowledge about words used in text. The first option was to use electronic dictionaries that would classify words by attributes such as type and subject matter. This was the state-of-the-art but relied on humans to compile these dictionaries and trusted them to be consistent and thorough in collecting and defining the words.

I saw a second option which was to teach the machine to discover the context of a word based only on the way that it is used in relation to other words. This idea was inspired by my second role of a dictionary (described above) and would eliminate the dependence on external sources of knowledge about words. The next step was to start evaluating existing approaches while I developed my own.

The full text of all publications cited in this article (except for the OED) can be found on either Project Gutenberg or the Internet Archive and read (or listened to) using Skimcast; that is how I did the research for this post.

IMAGE: Taken from Noah Webster’s “An American Dictionary Of The English Language” published in 1828. Scan courtesy of the Internet Archive.

6 Replies to “Words And Their Meanings And The Role Of A Dictionary”

  1. Bill, your project is changing life as we have known it. It is wildly exciting to discover the new possibilities available through your work. May God bless your efforts and may your rewards be manifold.

  2. Bill- thanks for cracking the door on a solution to a very complicated but needed solution. What’s so exciting is how varied and helpful what you are doing is to everyone- sight impaired or not. I feel privileged to get to read about it. Thanks for using your gifts in such a thoughtful and meaningful way.

    1. Thanks for your encouragement! I am excited every time I hear about someone else being helped by it. Language is something that we all share and so I love being able to write about it.

    1. Thank you, Jill! It was great to meet you and I look forward to working with you. Thanks for the suggestions!

Leave a Reply

Your email address will not be published. Required fields are marked *

Download Skimcast!