When my best childhood friend grew up, he decided to become an archaeologist. During his graduate training, he was in charge of a dig in the Mississippi river valley which unearthed an odd structure. In the midst of the native people’s dwellings, was found a circular enclosure made of closely spaced wooden posts and containing a large pile of turtle shells. The undergraduates were eager to speculate about the purpose of the structure, but my friend cautioned them against it.
“We can’t be sure of its use,” he said”, and we can’t just guess based on what we might use an enclosure like that for today. We can’t just assume they were running a turtle ranch here. Why would they do that with a river full of turtles just a quarter-mile away? We have to put it in context of the surrounding village and the environment of the time, look for other examples and see if there are any modern structural analogs. Then we can make a guess, but it will still just be a guess.”
The next day the professor in charge of the dig came around on a rare site visit to see how things were proceeding. The students were eager to show him the mysterious ring of posts with its pile of shells.
Upon seeing their find, the professor remarked without hesitation, “Huh, must have been a turtle pen,” and promptly resumed his walking tour of the dig.
I don’t know if archaeology has an excuse for this kind of thinking, but medicine does:
Life is short. The art is long. Experience is difficult.
We can be forgiven for resorting to teleological assumptions now and again in medicine. With limited time and incomplete information, we must sometimes act on hypotheses which attribute function to structure and purpose to processes. Lucky for us, there’s plenty of slop in the system, so even if we’re wrong at the start, we usually get a second chance. We are trying to get away from teleology, though. “Evidence based medicine” and “scientific medicine” are the names that we have given that effort.
We are trying to get away from teleology because we have been burned by it. We thought that the body made pus to fight off bacterial infections, so for years, when we saw people with respiratory illness cough up phlegm with pus in it, we gave them antibacterial medications. We were wrong, not just about the purpose of pus, but in attributing a purpose to pus. Again, it was an understandable mistake, given the long history of debate regarding the merits of pus. Was it a good sign, or a bad one? Should we encourage or discourage its formation? It turns out we shouldn’t have been focusing on the pus at all, but on the outcome of our purposeful intervention in the underlying process that produces the pus.
Purposeful results and final causes apply prospectively to human endeavors alone, and even there it’s often difficult to tell whether, when our actions are associated with the desired result, the outcome is due to our actions or simply due to fortuitous circumstances. Applied retrospectively or to processes and structures beyond our control, teleology is a sure mistake.
When we assign an endpoint to a process, we presume causation and correlation must be proven. Humans are notoriously bad at that. In systems which we can’t duplicate or control, we can always tell a causal story (I’m looking at you evolutionary psychology, intelligent design, cosmological fine tuning). But those stories are just interesting rationalizations, sharing the merits of a fairy tale in that they reveal more about us than the subject matter. Our fairy tales are harmless when they are about the universe, the origin of life, turtle ranches or anything else beyond our control. When we tell teleological stories about processes we do seek to influence (and can) we court tragedy.
The practice of bleeding was based on one such tale: the story of homeostasis. We still tell it today, but we tell it as metaphor instead of fact. The story is based on the simple observation that, when a person becomes ill, they go through a series of changes in their physical state which ultimately ends in either the restoration of their previous state, or death. Having observed other systems, the Greeks thought that the process of illness looked like a disequilibrium. Having observed associated changes in fluids which emanated from the body, they attributed the disequilibrium to an imbalance in those fluids. We can hardly blame them for the limits of their observations. We can’t fault their hypothesis. However, we can fault their method.
They didn’t just postulate an imbalance in the humors as a cause of illness, they presumed a balance of the humors as a state the body sought. The difference in these two points of view is subtle, but crucial. If the balance of fluids is seen as descriptive then restoring health by balancing the fluids remains a working hypothesis. It admits that other factors may determine the observed equilibrium. It leaves open the possibility that the observed flux of humors is a secondary phenomenon. Most important, it leaves physiologic equilibrium as a simple description, instead of presuming that it is a purpose with causal powers.
Given a description and a working hypothesis, physicians would look at their efforts to balance a patient’s humors with a critical eye. As a teleological assumption, with equilibrium as a “final cause” under Aristotle’s system, the idea creates an entirely different viewpoint. With humoral balance rooted in the body’s design, variances in expected observations must be due to inadequate methods or incomplete knowledge of the humors. For this version of the “balancing the humors” hypothesis, failure is not an option.
Now, the ancient Greeks may have weathered this kind of assumption better than their heirs. They loved to fight with each other. In the face of inconsistent outcomes from humor-balancing interventions, they were likely to call Aristotle and Hippocrates idiots or just ignore the under-girding theory of causes altogether in favor of their own pet theory. Definitive statements naturally took a healthy beating in the Greeks’ intellectual environment. The Romans, and the Europeans who came after them, were much more pious.
As a result, no one questioned the teleological assumption, out of reverence for its sources, and the vital fluids persisted in medical thought owing largely to the idea of homeostasis by design. No matter how apparent the flaws in our understanding of the blood, bile and phlegm, they were somehow attached to the homeostatic goal of the body. As long as physicians saw that equilibrium as the body’s goal, they could reconcile any discrepant observations with the over-arching story and persist in practices such as bleeding. It fell to investigators outside of the medical profession to discover the secondary nature of the humors. Only then did the practices aimed at balancing the fluids truly begin to fade.
But long after bleeding and the balance of fluids fell by the wayside, the tale of homeostatic purpose continued to plague medical science. Physicians continued to view physiology as directed toward an end. For example the heart was seen not to pump blood, but to be a pump. Therefore, medical students were instructed to never administer medications called beta-blockers to patients with heart failure.
Beta-blockers stick to proteins in the membranes of heart cells called beta receptors, which normally bind adrenaline. Via the beta receptor proteins, adrenaline stimulates the heart to pump faster and with more force. In heart failure, the heart can’t contract forcefully or fast enough to keep up with the volume of blood returning to it from the veins. If the heart is a purpose-built pump, beta blockers should be anathema in the setting of heart failure. But in reality, when given to stabilized heart failure patients, beta blockers reduce long-term mortality by about one-third.
We don’t yet know exactly how these medicines achieve such a feat. We do know why they are not inevitably detrimental in heart failure. It is because the heart pumps, but it is not a purpose-built pump. The heart is instead a group of cells which inhabits a specialized niche in a system of many cells all with complimentary and competing characteristics, existing in a state of equilibrium which, in deference to tradition, we call homeostasis.
Our physiology doesn’t try to maintain homeostasis any more than erosion tries to form a natural arch. The arch forms (rather than crumbling like the sides of a stream-bed) because it is geometrically stable given the geology. The arch persists because it is geometrically stable, and so we frequently see natural arches where the climate and geology allow. Nobody marvels at this, speculating about a conspiracy between sandstone and weather patterns. Then again, few people have an emotional stake in natural arches. The same is true of our physiology, minus the low stakes. There is no overall homeostasis sensor or hormone in the body. There is no homeostasis conspiracy.
So, we have abandoned the notion of purpose in physiology, and that simple maneuver has allowed us to discover things like the survival benefit which beta blockers produce in heart failure. This move is the principle behind the randomized, controlled clinical trial. All along, it wasn’t ignorance holding us back, but the project of rationalizing our knowledge to traditionally understood, teleological models.
Of course, the questions driving evidence based medicine don’t start from nowhere. Scientific medicine asks questions based on the results of previous investigations and hypotheses derived from basic science discoveries regarding the components of physiology and their relationships. Some of these hypotheses are even most easily stated in terms of purpose. But those statements are now understood as metaphor, rather than bare fact.
Beyond the fecundity of this change in method, the move away from teleology finally brings some redemption for poor Hippocrates. Rather than using it as an excuse, we can understand his aphorism, “Life is short. The art is long. Experience is difficult “, properly again – as an admonition about method. Be skeptical. Remember that your viewpoint is limited. Watch out for overarching narratives. Good advice, and not just for medicine, but for all those turtle-ranch theorists out there (I’m looking at you intelligent design, cosmological fine tuning, evolutionary psychology…).