How Dewey Decimal Classification Works
Have you ever come into a public library and searched for a book in the nonfiction section? If so, you probably encountered a confusing array of letters or numbers (or possibly both) on the spine of the book. These are called classification schema, systems that set a framework within which a library can work to order books in an organized, uniform way that is easily transposable between different settings.
You may see a series of letters and numbers such as A420.H36 (top picture, left). This is an example of Library of Congress Classification. On the other hand, you may see only a number, such as 822.33. This is an example of Dewey Decimal Classification (DDC for short), the system used by the Mercer County Library System (bottom picture, left). In this blog post I will discuss the history and functionality of the Dewey Decimal System of Classification.
A Brief History of DDC
DDC was developed by Melvil Dewey in the late 1800’s while he was working at the Amherst College Library. Dewey managed to create a system that he felt would best address the needs at the College, published in the “Decimal Classification and Relativ Index for arranging, cataloging, and indexing public and private libraries and for pamflets, clippings, notes, scrap books, index rerums, etc.” [sic]. He publicized it in Library Journal, and as the benefits became clear, adoption took off. Today the system is in its 23rd edition (published in a document with a much shorter title). It is regularly modified based on input from librarians around the country and is the most widely used system of library classification.
Hierarchical Structure
One of the major innovations of the DDC method of classification was that it arranged books relative to each other based on topic. Prior to DDC, libraries arranged books based on fixed positioning, where a book would keep the same position on the shelf based entirely on its physical size and date of acquisition. This may seem obvious in retrospect, but DDC is far superior. Patrons are able to shelf browse, or find a single location in the library and have all of the books on a single topic easily accessible.
DDC manages this by using base 10. Information is divided into subsets of ten to create a three-digit number followed by any number of digits after the decimal, depending on how specific the user wishes to be. There are ten main classes, each of these have ten subclasses, and so on.
For example, consider a book about the New Testament. Of the top ten main classes, this fits best in the 200 class, for religion. Within 200 I would select the 220 subclass, for information about the Bible, and within that class I would choose the 225 division for New Testament. It is really that easy.
Going Beyond the Hierarchy: Built Numbers
Of course, not all information can fit easily within a structurally defined number like in the example above. We only have 1000 whole numbers to choose from, and of those only 908 are actually defined, meaning that there are even fewer divisions to use.
Therefore, the DDC manual has devised a unique system of what are called “built numbers,” which gives far more flexibility than Library of Congress Classification. Built numbers define specific rules for creating a number that is unambiguous no matter the situation, but can be far more specific than the divisions already defined.
For example, suppose I want to classify a book that is about a Super Bowl game in Philadelphia. First, I would find the most specific defined number that fits this, in this case 796.332648. But there is no number defined for Philadelphia, so what the DDC schedule does is create a set of rules for making the number more specific, based on what class you are using. Accessing a specific table of subdivision rules, I can add 09 to the end of the number to say that I am treating this subject based on geography and then extend the number based on the where of the manuscript’s primary topic. You can extend this as deep as you want, based on how specific you want to be. The complete number for a Super Bowl in Philadelphia is 23 digits.
This can be done with any topic, not just geography, based on the rules in that class for adding subdivisions. This flexibility allows librarians to classify books based on nearly any topic and ensure that in large collections, books will not share similar or identical numbers.
I hope that this brief explanation can help you, as a patron of the library, understand exactly what you are looking at when you see one of those numbers on the shelf. If you are interested in more information, you can find books on library science under the 020 subclass of the 000 Class for Computer Science, Information, and General Works or search the library catalog.
DDC was developed by Melvil Dewey in the late 1800’s while he was working at the Amherst College Library. Dewey managed to create a system that he felt would best address the needs at the College, published in the “Decimal Classification and Relativ Index for arranging, cataloging, and indexing public and private libraries and for pamflets, clippings, notes, scrap books, index rerums, etc.” [sic]. He publicized it in Library Journal, and as the benefits became clear, adoption took off. Today the system is in its 23rd edition (published in a document with a much shorter title). It is regularly modified based on input from librarians around the country and is the most widely used system of library classification.
Hierarchical Structure
One of the major innovations of the DDC method of classification was that it arranged books relative to each other based on topic. Prior to DDC, libraries arranged books based on fixed positioning, where a book would keep the same position on the shelf based entirely on its physical size and date of acquisition. This may seem obvious in retrospect, but DDC is far superior. Patrons are able to shelf browse, or find a single location in the library and have all of the books on a single topic easily accessible.
DDC manages this by using base 10. Information is divided into subsets of ten to create a three-digit number followed by any number of digits after the decimal, depending on how specific the user wishes to be. There are ten main classes, each of these have ten subclasses, and so on.
For example, consider a book about the New Testament. Of the top ten main classes, this fits best in the 200 class, for religion. Within 200 I would select the 220 subclass, for information about the Bible, and within that class I would choose the 225 division for New Testament. It is really that easy.
Going Beyond the Hierarchy: Built Numbers
Of course, not all information can fit easily within a structurally defined number like in the example above. We only have 1000 whole numbers to choose from, and of those only 908 are actually defined, meaning that there are even fewer divisions to use.
Therefore, the DDC manual has devised a unique system of what are called “built numbers,” which gives far more flexibility than Library of Congress Classification. Built numbers define specific rules for creating a number that is unambiguous no matter the situation, but can be far more specific than the divisions already defined.
For example, suppose I want to classify a book that is about a Super Bowl game in Philadelphia. First, I would find the most specific defined number that fits this, in this case 796.332648. But there is no number defined for Philadelphia, so what the DDC schedule does is create a set of rules for making the number more specific, based on what class you are using. Accessing a specific table of subdivision rules, I can add 09 to the end of the number to say that I am treating this subject based on geography and then extend the number based on the where of the manuscript’s primary topic. You can extend this as deep as you want, based on how specific you want to be. The complete number for a Super Bowl in Philadelphia is 23 digits.
This can be done with any topic, not just geography, based on the rules in that class for adding subdivisions. This flexibility allows librarians to classify books based on nearly any topic and ensure that in large collections, books will not share similar or identical numbers.
I hope that this brief explanation can help you, as a patron of the library, understand exactly what you are looking at when you see one of those numbers on the shelf. If you are interested in more information, you can find books on library science under the 020 subclass of the 000 Class for Computer Science, Information, and General Works or search the library catalog.
- Ross H.
Top photo courtesy of Wikipedia.
Comments
Post a Comment