I’ve noticed a tendency for bureaucratic procedures to become more convoluted and opaque over time. I think this happens much more quickly when there are groups with conflicting interests working on a common project, and neither has much ability to compel the other to do something. If the process is governed by rules, it is natural to appeal to the rules to justify one’s preferred course of action. However, motivated people can find or simply invent ambiguities, loopholes, etc. So you might have a series of interactions that looks roughly like this:

Group 1: I want to do X.

Group 2: Rule A says we shouldn’t do X.

Group 1: But the wording in this sentence of rule A is open to differing interpretations, and under my interpretation X is allowed.

Group 2: Your interpretation is incorrect. Here are other regulations, and other parts of the same regulation, that clearly establish the intent and correct interpretation of rule A.

Group 1: [Proceeds to apply the same process of finding ambiguities to the other regulations appealed to.]

Group 2: [Continues knocking down group 1’s appeals to ambiguity.]

Unless someone has the power to resolve the dispute by fiat, there’s no natural end point to this process. Supposing group 1 is incorrect, there is no point at which that mere fact forces group 1 to abandon its position. So, in practice, you probably end up with one of two outcomes: decision by attrition, in which whoever persists longest and most vehemently wins; movement to a higher level of abstraction. In the latter case, one might reason, “Well, clearly our two groups have some disagreement about how the process works, and we need to resolve that before we can work through the logjam on any particular case. Let’s have a a few people from each side meet for the next few months and write up a process we’ll all agree to follow in future disagreements of this type.” Unfortunately, this increases the pile of policy material that both sides can appeal to in looking for and trying to close loopholes, and makes it more difficult for a neutral party to read the policy material and understand how the process is actually supposed to work.

Another possibility is to try to exert power by adding steps to the process. If you want to say, “No,” but can’t, instead you say, “Yes, and here are three forms to fill out,” and make sure each of the forms is unclear and requires input from multiple parties who don’t normally work with each other.

Although I think this tendency is much more dramatic in bureaucracies within which there are opposing interests and little direct power, I think it’s fairly universal to some extent. The example that prompted this post is a new combination published by Rupert Barneby in The Great Basin Naturalist in 1986:

“Lotus plebeius (T. Brandg.) Barneby [based on: Hosackia plebeia T. Brandg., Proc. Calif Acad. Sci. II, 2: 133 (Brandegee s.n. 30 Apr 1889, “El Rancho Viejo” [near Calmalli, Baja California Nort, +/- 28° 15’ N]].”

For those not familiar with botanical nomenclature, Brandegee published the name “Hosackia plebeia” and Barneby is moving that name from the genus Hosackia to the genus Lotus. There is no further discussion preceding or following the text quoted above, that is the entire thing. In this case, I was hoping for a little more information. Some botanists who have worked on these plants recognize Lotus plebeius as a species. Some do not, and include the name as a synonym under another species. We can infer that Barneby believes Lotus plebeius to be a distinct species—why would he create the new combination “Lotus plebeius” unless he intended to use it? if considered a synonym, one can leave it be in Hosackia—but we have no idea why. Nor why he thinks it should be in Lotus, for that matter. So, in this case, I’m on the side of wanting a bit more complication in the process. If you’re publishing a new combination, give us a few sentences of explanation.

In creating the self-published Journal of Semiarid Environments, though, I’m intentionally nudging the process away from procedural complication. For those used to the current process in academia, this can seem like lowering scientific standards. Science means peer review, and publishing in real journals! If we let people publish whatever they feel like, it’ll be anarchy! (I exaggerate, but you get the idea.) What interests me about Barneby’s publication of Lotus plebeius is that it’s nothing but procedure. It does precisely the minimum that is procedurally necessary and no fault can be found with it. It just isn’t very helpful to readers seeking to increase their understanding of the plants involved–which is, after all, the greater purpose that botanical nomenclature exists to serve! What’s missing is the part that is not mere procedure. Trying to resolve that deficiency and produce botanical insight by making the rules more complicated is like creating a new form for someone to fill out, or having a committee work on increasing the pile of policy documents.

Attention directed at procedure is attention directed away from one’s purpose. The best procedure is the one that minimizes that redirection of attention.

In bureaucracy, excess attention to procedure is probably an indication that the organization’s structure prevents people from working productively to fulfill their responsibilities. The organization is broken in a way that can’t be solved by paying more attention to procedure. Paying more attention to procedure is both a symptom of and a reinforcement of the organization’s failure.