Some of the most cited criticisms against Bloom’s taxonomy were written by Brenda Sugrue, PhD. She presents three main criticisms and two alternatives to Bloom. However, her criticisms are problematic.
Sugrue’s first criticism is that Bloom’s taxonomy is invalid, citing that it is almost 50 years old (at the time of her criticism). Of course, reversing the argument from antiquity would still be a logical fallacy, as being old doesn’t automatically make something invalid. Unfortunately, she seems to be using its age as a reason for its invalidity in itself. She also claims that it is unsupported by any research on learning. Though she wrote this criticism in 2002, she seems to have missed the fact that there was a major revision of the taxonomy in 2001 by Anderson, et al. The new Bloom’s taxonomy takes more advantage of dozens of years of scientific research on learning.
Her second criticism is that it is unreliable. Sugrue argues that it cannot be applied consistently by different educators. This criticism misses the fact that all of teaching cannot be applied consistently by different educators. If every teacher did the exact same things in the exact same ways, students would struggle. Different students and different situations require different approaches. Teachers cannot be consistent across all domains.
Sugrue’s final criticism is that Bloom’s taxonomy is impractical. She brings up the same point twice (under both the previous criticism and this one) that the whole taxonomy could be reduced to two levels. All of the “higher level” categories are often lumped together, and “knowledge” accounts for all “low order thinking.” This strikes me as an example of the continuum fallacy. Just because one may be unable to draw a specific line to distinguish different parts of a continuum does not mean that those distinctions do not exist. The lines between the categories on Bloom’s taxonomy are fuzzy, but just because one could lump things together does not mean that one has to. Bloom’s levels can be seen as overlapping circles on a Venn diagram, each has its own domain but there is overlap between them. Lumping all the “higher levels” together just destroys the usefulness the taxonomy previously had.
After giving criticisms, Sugrue presents two possible alternatives to using Bloom’s taxonomy, neither of which are particularly impressive. The first is the Content-by-Performance approach, which is a simple matrix of 5 types of content (facts, concepts, principles, procedures, and processes) and 2 levels of performance (remember and use). I don’t think this approach is inherently bad, and it can certainly serve as a useful framework for curriculum planning (much like Bloom’s), but it has the exact same problems she criticized of Bloom’s taxonomy. One could, very easily, lump all the contents together as being “facts” because memorizing the steps in a process is not any different than memorizing other facts about a process. The level distinctions can also easily disappear, as “recalling a fact” and “recalling a fact during a task performance” are not necessarily the most useful distinctions to make. This approach has as much invalidity, unreliability, and impracticality as she argues Bloom’s does.
Finally, she presents another alternative which is to not use any taxonomy at all. One could make this argument for any taxonomy that humans use in any subject. A taxonomy is not strictly necessary, but it is useful. The reason people created Bloom’s taxonomy to begin with was to help organize skills and abilities that teachers need to consider when planning how to teach. No teacher needs to use them, but having a framework is extremely valuable and this alternative isn’t much of an alternative. It is a bit like saying that hammers do a poor job of fastening certain kinds of things together, so you should just not use any hammers at all.
Use it like any other tool: for the specific purposes and occasions that it was meant for.