[Originally published as What are Lumpers and Splitters in Taxonomy?]
Lumpers and Splitters: Individuals or methodologies that broadly or narrowly group together similar members into a species (or other taxonomic rank).
Taxonomy: the branch of science that classifies living organisms into groups.
Defining Lumpers and Splitters
Within the realm of taxonomy, there have always been those who are known as lumpers and others as splitters. Those who are lumpers take many similar organisms and group them together as a single species. Splitters will take these same organisms and classify them as multiple species.
While this may not sound like a significant issue, when put into practice it can have some dramatic effects. I remember, about 20 years ago, when a suggestion was made to split many of the dragonfly species in Ohio into multiples.
The effect would have greatly increased the number of endangered species and the amount of local lands needed for their protection.
Hybridization* as a Splitter
Within the realm of baraminology the same effect occurs, but it is based more on the methodology that is used to determine created kinds. Those who determine kinds based on their ability to hybridize and the more classical methods of taxonomy tend to be splitters.
Statistics* as a Lumper
On the other hand, statistical baraminology has thus far shown itself to be a lumper by grouping together many more species, genera, or families than other methods allow.
Lots of Research Still to Come
Both using hybridization and statistics as methods to group creatures only lead us to educated guesses. While successful hybridization gives us a positive connection between species as members of a single kind, all the other forms of connections are necessarily tentative.
But, statistical comparisons could prove useful to help point researchers in the directions they need to find data to make more accurate connections and discover the limits of a kind.
Let’s Try This with Turtles
A good example of this process can be seen in determining turtle baramin. Evolutionary taxonomy classifies 14 families of turtles. A study by Timothy Brophy, Wayne Frair, and Darlene Clark used hybridization and traditional techniques to determine a few connections and settled with eleven turtle kinds. A statistical study of turtles made by Todd Wood showed only 5 kinds of turtles. Interestingly, the larger study did not find any evidence to falsify the study by Wood, but at the same time, did not have enough evidence to support it either.
For my work I favor the traditional splitting method while saving statistical studies for when a significant difference is found. I expecte we will collect more data over the coming decades, bringing more decisive results and a more accurate number of baramin.
[Editor’s note: A definition of using hybridization as a classification tool from Answers in Genesis:
For living organisms, hybridization is a key criterion. If two animals can produce a hybrid, then they are considered to be of the same kind.1 However, the inability to produce offspring does not necessarily rule out that the animals are of the same kind, since this may be the result of mutations (since the Fall).