The Rule of Three

Section 5.7 (D) of the DDC Introduction presents the rule of three:  "Class a work on three or more subjects that are all subdivisions of a broader subject in the first higher number that includes them all (unless one subject is treated more fully than the others)."

Let's take Dunham’s History of Denmark, Sweden, and Norway as an example.  Denmark is classed in 948.9 Denmark and Finland, Sweden is classed in 948.5 Sweden, and Norway is classed in 948.1 Norway.  By the rule of three, this work is classed in the first higher number that includes them all, i.e., 948 Scandinavia and Finland. 

Given that Scandinavia consists of Norway, Sweden, and Denmark, the rule of three simply confirms what most classifiers probably would have done intuitively.  The rule becomes more useful when the subject of the work consists of the equal treatment of three or more topics of a rather larger set of subtopics.  For example, consider a work on what at one time were called the "outer planets," Uranus, Neptune & Pluto.  Individually the three topics would be classed in 523.47 Uranus, 523.48  Neptune, and 523.4922  Pluto.  The first higher number that includes them all is 523.4 Planets, asteroids, trans-Neptunian objects of solar system.

Does the rule of three concern only notational hierarchy, or does it also concern structural hierarchy?   On the one hand, the way the rule of three is stated—especially with its mention of "the first higher number that includes [all of the subdivisions]"—suggests that it concerns notational hierarchy only.  On the other hand, if the rule of three were to interact with the comprehensive number for a topic, interpreting it as applying to structural hierarchy would generally lead to appropriate classification results.  (Alternatively, we could say that notes that give the comprehensive number for a subject override the rule of three.  In any case, we get the same results.)

To explore this topic further, let’s consider the following segment of the classification scheme:

224.6     *Hosea (Osee)

224.7     *Joel

224.8     *Amos

224.9     *Minor Prophets

For Hosea, see 224.6; for Joel, see 224.7; for Amos, see 224.8

224.91           *Obadiah (Abdias)

224.92           *Jonah (Jonas)

224.93           *Micah (Micheas)

. . .

The Minor Prophets are twelve in number; nine are listed in the direct subdivisions of 224.9 Minor Prophets, while the other three are given in coordinate notation at 224.6–224.8, but subordinated to 224.9 through a series of see references.  The rule of three calls for a work on Obadiah, Jonah, and Micah to be classed in 224.9 as the first higher subdivision that includes them all.  If a work on three of the twelve Minor Prophets in direct subdivisions of 224.9 should be classed there, it would also make sense for a work on Hosea, Joel, Amos, Obadiah, Jonah, and Micah—that is, a work on six of the twelve Minor Prophets, three of which are in direct subdivisions of 224.9 and three of which are subordinated to 224.9 through see references—to be classed in 224.9 also, as the comprehensive number for Minor Prophets.  By the same token, we also see a work on Hosea, Joel, and Amos being classed in 224.9, as the comprehensive number for Minor Prophets, even though these three Minor Prophets are "only" in the structural hierarchy under 224.9 and not in its notational hierarchy.   (By the way, such works are highly likely to be commentaries, to be classed in 224.907 Commentaries on Minor Prophets, but the building of that number gets into issues that would sidetrack us.) 

A work on the history of Denmark, Sweden, and Norway should be discoverable by a search on the history of any of the three countries; a work on Uranus, Neptune, and Pluto should be discoverable by a search for any of the three astronomical bodies; and a work on Hosea, Joel, and Amos should be discoverable by a search on any of the three Biblical books.  Statements here about the number that the rule of three instructs us to use should be understood to address the assignment of the standard DDC number, captured in the 082 field (Dewey Decimal Classification Number) of the MARC bibliographic record.  The numbers not chosen as the standard number should be considered for inclusion in the 083 field (Additional Dewey Decimal Classification Number).