Scrum and Empiricism

June 26, 2020

“Scrum is founded on empirical process control theory, or empiricism.”

What? Empiricism?

The first time I read this sentence in the Scrum Guide, I was quite impressed. As a self-professed pseudo-philosopher, I had a vague understanding of philosophical questions on which empiricists and rationalists generally disagree with each other. It’s an ongoing argument in the history of human thought. To confuse the matters further, technically you can be a rationalist in one area (say mathematics) and an empiricist in another area of physical sciences. But the dispute generally holds true.

The fact that Scrum takes inspiration from empiricism tells us something important about it.

The question underlying the philosophical debate mentioned above is this: How can we gain knowledge? What is the ultimate source of our knowledge?

How would you answer that question?

If you think your primary source of knowledge is your sense perception, then I congratulate you for being a child of a Scientific age that relies heavily on empiricism to learn about the world. The Scientific Method, (if there really is such a thing) relies on sense perception. I would argue that in modern societies the need for “evidence” to make any claims to knowledge is so pervasive that we can’t imagine any other possibility. The rationalist positions would sound silly at first, unless we think very carefully about rationalists’ view of how knowledge is achieved.

An empiricist would believe that all our knowledge is derived from our senses. We don’t have any innate knowledge. Now this has worked well for Science in the last 500 years or so. Scientists like reproducibility of their results. In fact, recently there was a story published on BBC about a “reproducibility crisis” in the scientific world.

“Replication is supposed to be a hallmark of scientific integrity”, the above article tells us.

Rationalists, on the other hand hold that at least some of our knowledge is derived from reason alone. There is a wide range of opinions among rationalists regarding the relative importance of reason for acquisition of knowledge. Plato is probably one of the earliest philosophers showing rationalist tendencies. His whole idea of eternal Forms and doctrine of knowledge by recollection have strong smells of rationalism as it was developed subsequently. I guess what unites rationalists is their refusal to treat sense data as the only foundation for knowledge acquisition.

And that’s exactly the problem I have with empiricism: exclusivity. Empiricists don’t leave any chance of any other source of knowledge. But I do agree that Science totally relies on empiricism. Being scientific about something means being an empiricist.

And that makes Scrum a scientific enterprise.

And don’t get me wrong: it’s a good thing. Scrum is all about inspecting and adapting. As the Scrum guide tells us, each event in Scrum is a formal opportunity to inspect and adapt something. If findings of a team are backed by evidence and observation, we have something tangible to talk about.

If, for example, we observe that not having an automated CI/CD setup adversely affects team performance, and if it is an observation that is shared among multiple team members, we can then do something about it.

Empiricism makes it possible to have a conversation. It takes away opinions that can’t be substantiated with evidence.

In my opinion, that is a huge contributor to the success of Scrum as a framework.