Good Explanations as a Foundation for Critical Thinking
The Deutsch Hard to Vary Standard
“I argue that all progress, both theoretical and practical, has resulted from a single human activity: the quest for what I call good explanations.” — David Deutsch
What is an explanation, and what is a good explanation? Simple questions that are not easy to answer.
Short answer: An explanation is a statement or account that makes something clear.
Longer: An explanation is a statement or series of statements designed to clarify the meaning, cause, or purpose of something. It serves to offer a logical and coherent narrative of an event, phenomenon, or concept, thereby rendering it more comprehensible. The goal of an explanation is to uncover the underlying relationships, causes, or mechanisms responsible for a given occurrence. Essentially, an explanation seeks to address the "How?" or "Why?" concerning a specific phenomenon.
The late evolutionist Daniel Dennett differentiated between two types of "whys" ("Darwin's Dangerous Idea: Evolution and the Meanings of Life"):
"what for," which refers to a teleological perspective, which addresses the question of what something is for, what its purpose or function is. and
"how come," which refers to a process narrative of how things came to be the way it is.
The quantum physicist and author David Deutsch, sees explanation as a statement that delineates how the world exists, behaves, and the reasons behind it. The 'why' aspect often entails interpreting visible phenomena through the lens of the unseen, conjecturing the existence of elements we cannot directly observe. This process can sometimes alter our perception of what we consider to be known. According to Deutsch, while this defines an explanation, many explanations fail to adequately fulfill this purpose.
As for the second question, what a good explanation is, we find Deutsch’s ideas, particularly his criteria “hard to vary” most interesting. This is an alternative to the similar idea we looked earlier: Abduction, which some define as inference to the best explanation (but disputed by others) . Both Abduction and good explanation are about making plausible hypotheses based on available data or knowledge, and selecting the best ones.
Deutsch sets high standards for what constitutes a good explanation. The criteria include:
Hard to Vary: A good explanation should be difficult to change without losing its essence. He writes, "The explanation should be hard to vary without losing its explanatory power." This means that if you can easily tweak the explanation and still have it fit the facts, it's probably not a deep or reliable explanation.
Example: Think about the theory of gravity. Newton's law of gravitation isn't just a guess that objects attract each other; it gives a precise mathematical relationship that predicts how objects move. You can't just change the formula a little bit and expect it to still work – it's "hard to vary."
Provides Deep Understanding: A good explanation doesn't just tell you what happens, but why it happens. It gives you insight into the underlying principles. Deep understanding implies an answer to Dennett’s “how come”, and typically enables more precise predictions.
Example: Instead of just saying "plants grow towards light," a good explanation would delve into the biological mechanisms of phototropism – how cells in the plant's stem elongate on the side away from the light, causing the plant to bend towards the light source.
Testable and Falsifiable: Deutsch is a big fan of Karl Popper, who argued that scientific theories should be testable and falsifiable. This means that a good explanation can be tested by experiments, and there should be a way to prove it wrong if it's not true.
Example: The theory that "all swans are white" is testable and falsifiable. You can look for swans and see if they are white. Finding a single black swan would falsify the theory.
Unification: A good explanation often brings together previously unrelated phenomena under a single framework. It shows how different observations are connected.
Example: Darwin's theory of evolution by natural selection is a classic case. It not only explains the diversity of life but also connects it to geology, paleontology, and even the distribution of species.
Here, we will concentrate on the first criterion, as the others have been extensively discussed. But all the above criteria are equally important and they characterize what good explanations are.
"Hard to vary" suggests a condition that is robust yet sub-optimal, encompassing all necessary and sufficient parameters, such that omitting any would result in the loss of its essence.
In optimization problems, a local maximum/minimum is a solution that is better than its immediate neighbors but not necessarily the global optimum. Similarly, a hard-to-vary explanation is an explanation that is difficult to improve or modify because it is "stuck" in a local maximum, and attempts to vary it may only lead to a degradation of the explanation.
David Deutsch’s idea of "hard to vary" explanations can be meaningfully connected to Stephen Jay Gould’s theory of punctuated equilibrium and Thomas Kuhn’s concept of paradigm shifts. Both Gould and Kuhn offer frameworks that emphasize the robustness and transformative nature of scientific theories and biological processes, which align with Deutsch’s criteria for good explanations.
Another aspect of David Deutsch’s idea of "hard to vary" explanations is their parsimonious nature. An effective explanation should clarify and streamline a complex phenomenon, achieving compressive optimality through its non-superfluous character.
Practical applications of good explanations.
While the concept of good explanations was initially linked to science, it is argued that this concept is relevant to every aspect of life, including art.
Picasso’s Guernica painting is an excellent example of the principle of hard to vary explanation applied to Art.
Explanations are narratives. Narratives help us make sense of the world. Effective sense-making significantly impacts our understanding, decision-making, problem-solving and learning processes. When those explanations are really good, they can even lead to predictions we can test out in the real world.
They serve as the primary instruments to counteract biases like confirmation bias and stereotyping, and to foster critical thinking, which is essential for escaping the confines of echo chambers.
“Before the scientific revolution, they believed that everything important, knowable, was already known, enshrined in ancient writings, institutions and in some genuinely useful rules of thumb -- which were, however, entrenched as dogmas, along with many falsehoods.” - David Deutsch (TED Talk 2009}
The quote above was about the era before the scientific revolution, but let us not believe that the situation has improved today. The narratives generated by politicians and businesses in the current era are far more subtle, nuanced and widespread, exploiting the facilities of the information age.
So, in a time where narratives shape truth and the powerful (both politically and economically) are making up the dominant stories, possessing the skill to distinguish what constitutes a good explanation is invaluable.