How do I distinguish good theories from bad theories?

“Eine richtige Theorie ist das Praktischste, was es gibt”, Friedrich W. Dörpfeld, Grundlinien einer Theorie des Lehrplan (1873)

Later this was popularized by Kurt Lewin as: “There is nothing so practical as a good theory” Field Theory in Social Science (1951)

Proposition: any kind of intentional activity requires a theory that says that:

with intention a, under conditions b it is good to specify that intention a with goal c and the best means to achieve c and so work towards a is to is to do d.

to put that in ordinary language in the form of an example: “if I want to design ‘a nice house’, and know the relevant conditions at work in the situation I am operating within (considerations of the client, other stakeholders such as the municipality, the considerations of site and program, the available materials and technologies, their cost, what they require for implementation, cultural and socio-spatial considerations, my own artistic and ideological aspirations with regard to architectural design and whom I want to be, and no doubt I have forgotten something) then I shall proceed by taking these considerations as my lead and deploy the best method of design that I know in order to achieve my overall aim. And once I believe I have achieved that aim, I shall evaluate my success against the criteria that are implicit or have been made explicit in the considerations that have lead me to the design.”

From this viewpoint a theory stands for a coherent and consistent narrative formed from the point of view of an intention (a) that leads to an inferential process of giving and asking for reasons regarding the consideration of conceptualized properties of things and their expected behaviour under conditions (b) with the explicit purpose of coming up with good ends (c) as well as good means (d) to achieve those ends on the basis of that narrative.

In order to make sure you have read the conditions at play in a situation properly, and so perform the necessary preparations for acting well with regard to the possibilities offered, theory needs an analytical understanding of how things work and a critical position with regard to how that matters to us in the context of a particular purpose or goal so that means and ends can be carefully aligned on a thorough understanding.

What makes a theory a good theory? And how can I distinguish a good theory from a bad one? The complexities this seemingly simple question will get us into are labyrinthine.

In universal terms such criteria are quite easy to formulate. Let’s set one up using broad brushstrokes:

“A good theory reflects the working of the world accurately and helps shape practice in such a way as to achieve good ends through good means.”

There you go, that wasn’t so hard. And the quotations I began this chapter with together form another good example. But when we look at each sentence a little more carefully we are really no further and ought to feel a little disappointed.

Explaining what ‘good’ means in concrete terms, (in terms of what I should want in this particular situation and how I should achieve it without nasty side-effects) is no easy task and yet that is really what we want to know. To use the word good is merely to say that you approve of something, but it does not say what you then need to do.

The word ‘good’ is what we call a universal. It is a concept that tries to group a particular set of properties, namely all things judged good. To say that something is good is to give it a form of approval. So the word good is used to put everything you approve of in the set of things that may be labelled as ‘good’. You can compare the concept to an empty vase that can be filled with a certain kind of thing but not with another. But the vase itself is only there to hold what is in it and we need to work with that, not with the vase.

In extremely stable environments where little changes from one situation to the next, and situations overlap in the conditions they display we can build up habits and traditions, knowing with reasonable certainty that the outcome will be what we expect of it. Even now, within our turbulent and changing world there are goods that are easily made concrete and which do not change unexpectedly: “if you are thirsty it is good to drink”. On this level there are innumerable unchanging goods we learn to wield in steering our course through life. But when things become more complex from our point of view, when conditions change, it can become very difficult to judge something good.

My thesis is that there is absolutely no short cut to judging theories good. And if you think there are, I worry for you.

What are the variables at work? To be able to judge a theory good we would want compelling and exhaustively complete answers to the following questions and have a method to align each of those considerations to each other and to our intention.

1. The considerations taken account of may or may not be the same as the considerations actually at work in a situation. Our understanding might be incomplete or faulty: Our frame of reference holding our concepts which we think with and on the basis of which we can make inferences may not hold sufficient concepts for a deeper understanding. And each of the concepts held may not be embellished with sufficient conceptualized properties to make proper inferential headway within the space of implications.
2. The above will affect the outcome of any inferential process, but so can the legitimacy of the inferences on the basis of the rules of logic and grammar as well as their conformance to empirical observation. It is possible to commit fallacies even with a full and well-structured frame of reference. It is also possible to reason correctly within the rules of logic whilst the outcome of that game of reasoning does not conform to empirical observation.
3. There are an indeterminate number of possible goals that can be formulated in any situation. Which of them are reasonable or sensible with reference to the interests of each of the stakeholders involved (both in the present and the future) depends on a point of view.
4. There is a discrepancy between the number of stakeholders you might find relevvant and the number of stakeholders that are potentially affected by your theory. The stakes may change as the situation in which each stakeholder finds themselves changes. Furthermore their stakes may conflict with each other. Then there are short term and long term interests. Indeed, the interests of each stakeholder relates to scale. In a particular situation, the interests of my arm (assuming that it wants to preserve itself, may not be the same as the interests of my whole body when under threat and these might diverge from what I call mine, such as my sister, friend, neighbour, car, tennis racket, community, club, nation.

So the variables we have to fit together in a happy way to judge a theory to be good are considerable, near infinite in fact, if we want to judge a theory good in an exhaustive way.

It is for this reason that we take the muddler’s perspective. In the end all we can do is make (design) decisions after we have reached reflective equilibrium, a point defined by John Rawls in any discursive exercise where we run up against the limits of our own knowledge and notice this by going off subject or by repeating ourselves or by simply not really getting any further. But do not equate this point with an objectively good theory. Such things simply do not exist except in our imagination, at least not from all possible perspectives as these soon start conflicting with each other.

The impossibility of achieving such a theory should not allow us to indulge in fatalism. (The thought that if we can’t achieve an objectively good theory we might as well give up)

If our task is to make ourselves, our offspring, and our environment flourish, then any progress achieved was worth pursuing.

In judging a theory we have only hope and experience. Experience can be properly analyzed and critiqued. Hope is tempered by experience and requires analogy to reach beyond it.

We may well think that new forms of intelligence can solve this complexity issue, and you may be right, but unless these new forms of intelligence can explain themselves and their reasoning, you will need to invest your trust in their doings. Which is another way of saying that you will be investing your autonomy in their authority. That is a gamble.

© jacob voorthuis, 2025. Please cite Jacob Voorthuis as the author, The Theoria Project as the title and the page address as the location. This work is licensed under a Creative Commons Attribution 4.0 International License. You are free to: Share — copy and redistribute the material in any medium or format Adapt — remix, transform, and build upon the material for any purpose, even commercially under the following terms: No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits. Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made.