Grades and Clades

Much of the confusion in textual criticism in recent years comes froma failure to distinguish genetic similarity from mere coincidenceof reading. Similar readings often imply kinship, to be sure -- but in aparticular case, it may be the result of accidental levelling. Instanceswhere this can happen especially easily include haplograpic errors,assimilation of parallels, and expansion of Christological titles.

The biological sciences have been aware of this problem for decades, andhave evolved terminology to deal with it. The term for a set of things(species, manuscripts) with a similar set of characteristics is a grade;the term for a set of things which evolved from each other is a clade.

To illustrate how this works in biology, consider this family tree ofthe reptiles and their descendents (as best I understand it, with someattempt to translate this into clearer jargon than the books I've read):

                                  proto-reptile                                        |                 ----------------------------------------------------------                 |                                            |           |           non-turtle-like                                turtle-like     |               reptiles                                    reptiles       |                 |                                            |     pelycosaurs            -----------------------                           |           |            |                     |                           |           |     dinosaur-like          lizard-like                       |           |       reptiles               reptiles                        |           |         |                        |                           |      therapsids    ------------            -----------------------           |           |    |          |            |          |          |           |           |crocodiles  dinosaurs     snakes   lizards    tuataras    turtles         |               |                                                          |             birds                                                     mammals

In viewing this, the key point is to realize that all living reptilesare descended from that firstproto-reptile, and no amphibians, fish, non-vertebrates, plants, or unicellularorgamisms aredescended from it. That proto-reptile is the ancestor only of living reptiles,mammals, and birds.

Note that, when we speak of reptiles today, we mean only a subset of thecreatures shown above:

                                  proto-reptile                                        |                 ----------------------------------------------------------                 |                                            |           |           non-turtle-like                                turtle-like     |               reptiles                                    reptiles       |                 |                                            |     pelycosaurs            -----------------------                           |           |            |                     |                           |           |     dinosaur-like          lizard-like                       |           |       reptiles               reptiles                        |           |         |                        |                           |      therapsids    ------------            -----------------------           |           |    |          |            |          |          |           |           |
    |          |            |          |          |           |  crocodiles  dinosaurs   snakes   lizards    tuataras    turtles                   |
     |           |           |      
               |                                                          |             birds                                                     mammals

This is based on a definition of reptiles as cold-blooded creatures which breatheair during their entire lives, do not secrete milk, do not have hair or feathers, andlay shelled eggs. Note that this is a definition based solely on characteristics.

Note also something rather absurd: If this genealogy is correct, modern birds aremore closely related to crocodiles, and indeed to snakes and lizards and tuataras,than they are to turtles and tortoises. And yet, turtles are "reptiles"and birds are not; they're listed as birds.

The word "reptile," then, is grade terminology.It is based on characteristics only, with no geneticor genealogical component. The word "amphibian" is also a grade;descendents of the first amphibians include not only frogs and salamanders andsuch (which we call amphibians) but also reptiles and mammals and birds (whichwe do not call amphibians).

A "clade" is entirely different: It consists of all creatures descendedfrom a common ancestor, no matter how similar or dissimilar they are. (The othertechnical term for this is monophyletic, but since it seems to refer to phyla --sort of the biological equivalent of text-types -- it's probably clearer to avoidthe term.)

If we wish to look at clades in the reptile diagram, the version below showsseveral (not all, of course)

                                  proto-reptile                                        |                 ------------------------------------------------------------
                 |                                                  non-turtle-like                                              reptiles                                                   |                                                   -----------------------
      |       turtle-like   reptiles      |      |
      |      pelycosaurs       |            |      |
            |                 |                dinosaur-like        reptiles         |    ------------    |          | crocodiles  dinosaurs                 |             birds
          |                 |                         lizard-like                      reptiles                            |                         -----------------------         |          |          |        snakes   lizards    tuataras
      |      |            |      |      |      |      |   turtles
      |           |           |           |           |        therapsids      |           |   mammals

Note that the leftmost clade shown, from dinosaur-like reptiles to birds, involvescreatures we call reptiles and creatures we call birds, and it involves both living andextinct species -- but all are descended from the ancestor of the group. By contrast,though snakes and crocodiles (say) are both reptiles, they are not descended from theancestor of croodiles and dinosaurs and birds, so they are not part of this clade.

There actually are clade names for a lot of the groupings above. All creaturesdescended from that first proto-reptile are "amniotes" -- so named becausethe eggs of reptiles contain an amniotic membrane to allow them to retain moisture whenlaid outside water. The group of all amniotes which are not mammals (that is, the turtles,birds, and what we now call "reptiles") are "sauropsids".The lizard-like reptiles are "lepidosaurs." Thenon-turtle-like reptiles, plus birds, are "diapsids." The group I've called"dinosaur-like reptiles" (crocodiles, dinosaurs, birds) are the "archosaurs."And the dinosaurs and birds, without crocodiles, are "ornithodires." Butwhen have you ever heard any of those names used?

One might think of clades as "vertical" relationships, from an ancestorto descentents. Grades are "horizontal" -- all the children in a school class,perhaps. They're all the same age, but they aren't from the same family.

There are several ways in which this distinction can be applied to manuscriptsand the history of the text. Manuscripts of, say, the ninth century are a grade butnot a clade; there is very little genetic relationship between, say, 33 and 461,but they are the "ninth century" grade. On the other hand, Dp,Dabs1, and Dabs2, while not of the same century, are clearlyrelated; since the latter two are copied from the first, they are unquestionablya clade.

Now consider a genealogy of the Versions of theNew Testament. (Please note that this is much simplified and very theoretical;not everyone agrees with this. Indeed, I don't agree with it myself. Butwe have to do something for demonstration purposes.)

           Greek Text               |  -----------------------------------------------------------  |           |               |         |        |          |Old Latin   Vulgate      Old Syriac   Coptic   Gothic    OC Slavonic              |               |                             |     ---------------          |                  --------------------     |             |          |                  |          |        | Anglo-Saxon  Old French   Armenian           Bulgarian   Russian  Serbian                              |                              |                              |                          Georgian

Now let's divide this into grades:

           Greek Text               |  -----------------------------------------------------------
   |           |               |         |        |          |               PRIMARYOld Latin   Vulgate      Old Syriac   Coptic   Gothic    OC Slavonic         VERSIONS              |               |                             |
     ---------------          |                  --------------------     |             |          |                  |          |        |       SECONDARY Anglo-Saxon  Old French   Armenian           Bulgarian   Russian  Serbian   VERSIONS                              |                              |
                              |                                              TERTIARY                          Georgian                                           VERSIONS

(In this context, I am using the terms primary version, secondaryversion, tertiary version not in terms of the importance of theversion but in terms of their relationship with the Greek text: A primaryversion is taken directly from the Greek, a secondary is taken from aprimary version rather than the Greek, and a tertiary is taken from asecondary version rather than from the Greek or a primary version.)

There can be, in some cases, some real use to this grade classification --for instance, if you are dealing with grammatical questions about the originaltext, a primary version is much more likely to carry useful information thana secondary or tertiary, simply because such distinctions will become progressivelymore blurred with each language the text is translated through. So thedistinction between primary and secondary versions is quite important for, say,the Hebrew Bible, where the versions are almost the only sources independentof the MT. But we don't need such help very much in connection with the GreekNew Testament, where we have so many independent witnesses.

Which means that grades of versions aren't much use in NT textualcriticism. The Old Church Slavonic may be a primary version in terms ofits origin -- but it is translated from the Byzantine text. Even if youare a believer in Byzantine priority, it's pretty useless, because thereare older and better Byzantine sources in Greek. On the other hand, theArmenian may be a secondary version (this is of course much disputed),but it is a secondary version from an interesting and early source. Thusthe Armenian version, though of an inferior grade, is of a superior clade.

If that is all that knowledge of grades and clades told us, it would be justa bit of minor terminology that we could all ignore. The true importance, however,lies in what we can learn from the biologists. The essential point is thatclades are much more useful than grades. Especially if you want to knowsomething about earlier ancestors. In an evolutionary context (which theprogressive change of manuscripts clearly is), grades are meaningful only ifall change is in the same direction. The chart above, for reptiles, showsthis. The grade "reptile" is meaningful if you think evolutionhad a goal, to get from amphibians to reptiles to mammals. But it is demonstrablethat this was not the goal; this is demonstrated by the fact that manythousands of unevolved reptile species still exist, and the reptiles alsoevolved into birds. Reptiles were not some sort of way station from amphibians tosomething "higher"; they aresimply a stage. To treat them as if they are something to progress pastis equivalent to saying that, because you drivethrough Baltimore on your way from Washington, D.C. to Philadelphia, thatBaltimore isn't a city and no one goes there!

And this matters because of -- text-types. (Youprobably knew all along that I'd get to this eventually.) Or, rather, attemptsto define text-types.

In recent years, I have been, frankly, disgusted by attempts to discusstext-types based on things like the Aland Categoriesand the Colwell-Tune Definition. Readingabout grades and clades finally made me realize what the problem is: All thesedefinitions are grade definitions! Colwell has a 70% agreement rule. The Alandcategories are basically categories of Byzantine-ness: Category I has almostno Byzantine influence, Category II is perhaps 50% conformed to the Byzantinetext, Category III up to 75% Byzantine, uncategorized manuscripts are less than90% Byzantine, and Category V are more than 90% Byzantine.

The results thus produced are as useful as anything based on grades can be.And it is much easier to determine membership in a grade than membership in aclade. But they are just grades; they are not clades. This means that theresults of these these classification schemes cannot be used genealogically. Ofcourse, many these days do not use genealogical methods. But it is importantto realize that these classifications simply aren't capable of being so used.

For more information on how biologists seek to find clades, see the articleon Cladistics.