Is it time to drain the entrepreneurship theory swamp?

The meaning of theory has been severely inflated, which has created an ever-growing mess rather than a consistent body of theory that improves our understanding.

A scholarly discipline is defined by its theory. In fact, one might argue that the theoretical explanations that are accumulated over time comprise the totality of our understanding for a certain subject matter or its phenomena. It is for this reason that scholarly contributions, notably the journal articles that have become the currency of our profession, are fundamentally theoretical. Without a contribution to theory, there is no contribution.

There are different ways of creating theory, of course. In the numerous editorials in our most highly regarded journals, senior scholars and editors share their view of what makes a theory and, importantly, a contribution to theory. Typically, this is done through applying “the scientific method” to formulate hypotheses that are then tested empirically on specifically collected data suitable for the task. But this is a practice that we borrowed from the natural sciences, arguably following economists’ “physics envy,” to make our discipline appear more “scientific.”

But the social sciences have always relied on a very different method for uncovering truths about the social world, which is arguably better suited for its dynamic, value-oriented, and behavioral, if not interpretive, subject matters: rationalism. We need theory to demarcate concepts and phenomena, untangle them, and produce explanations.

I mention this difference not as an argument for one or the other, only to illustrate that our theories differ not only in what they study and how they deal with observations. There are more fundamental (and potentially incompatible) differences too.

Add to this that the publishing requirement in our profession (read: “publish or perish”) pushes every one of us to produce numerous “contributions to theory” and to do so regularly. There is nothing wrong with the requirement, considering what theory is and does, but its institutionalization as a prerequisite for a position and currency in the career has made publishing - and thereby theory contributions - subject to Goodhart’s Law: “When a measure becomes a target, it ceases to be a good measure.”

The truth of Goodhart’s Law is behavioral: we measure something to gather knowledge about the state of the world. But when we use that metric as a goal or aim, we produce individual and collective incentives to meet the target. GDP is a good example. It was originally created by the economist Simon Kuznets as a means to approximate an economy’s ability to produce valuable goods, which itself approximates a population’s standard of living. But as GDP became the universal metric of national prosperity, it became a political goal. Consequently, what mattered was to effect higher GDP (the statistic) rather than what it intended to measure (prosperity). As a result, politics has a strong incentive, and thus finds ways, to improve the statistic whether or not it improves people’s standard of living. China’s GDP targets on every political level exemplify this. But it applies to most countries.

Goodhart’s Law in publishing has led to a greatly expanded view of what constitutes a theoretical contribution. To be honest, how many scholars today can explain the difference between a theory, a model, and a hypothesis? Or, to put it differently, what is the difference between theory and the “arrows” between the boxes in one’s paper? (Hint: they’re not the same.)

This theory “inflation” would not be a problem were it not for the fact that we, as a discipline, have also adopted the view that a theory contribution must add to theory. Clearly, this is not true. Einstein did not “add to” Newton. Marginalism in economics did not merely “add to” classical economics. These noteworthy contributions to theory in physics and economics, respectively (both of which so major that they caused Kuhnian paradigm shifts), constituted progress because they replaced prior theory.

In entrepreneurship research, however, we have adopted a faux humility that demands of us that we “build on” and “add to” the theories and explanations that we have accumulated over the decades. Do not point out flaws, challenge, or attempt to replace prior explanations. If you do, not only will your reviewers be the very people whose work you criticize (and therefore will mercilessly shoot it down), but you will not be considered a team player.

Whereas the psychology and institutional incentives are easy to explain, this practice is unsustainable and counterproductive. A discipline is defined by its theory, but what if its theories are inconsistent, build on contradictory assumptions, and are not empirically confirmed? What if physicists were publishing research that draws from whichever parts of Newtonian or Einsteinian physics fit with the thesis without regard for what is demonstrably false? What if economists produced analyses that picked whichever cherries deemed needed to publish from both the classical value theory (debunked in the 1870s) and marginalist theory (the standard since)?

Granted, entrepreneurship has not experienced much of a paradigm shift yet, especially not the type of revolution that both physics and economics have benefitted from (or suffered, depending on one’s perspective). But that is not the point. The issue is that we do not actually have a paradigm. We have a plethora of theories that we use side by side but that we have not bothered (for both good and bad reasons) to assess or evaluate. We certainly have not attempted (or even accept) to liberate our scholarly tradition from theories and speculations that are inconsistent or cannot be confirmed or validated.

I deliberately chose the word liberate to describe the weeding out of “bad” theory, because without the ability in research practice to assess, challenge, criticize, and debunk theories and findings, our body of theory not only retains those errors but builds on and thereby cements and augments them. This is highly problematic, because it raises questions about the reliability and value of the discipline’s entire scholarship. What is our contribution to the world?

A discipline is defined by its theory. Including those errors and misunderstandings that have not been weeded out. We need to find a way to rid our discipline of faulty, “bad,” and insufficiently explanatory theory. And build on and add to the rest.

We should drain our theory swamp, one might say. It would benefit the discipline.