As the line between nature and technology blurs, we need new ways to describe how the body works.
If you were a high school biology student in the United States any time in the last 30 years, chances are you would have been introduced to cell biology by a metaphor that compares cells to machines. In the Glencoe Science textbook The Dynamics of Life, which was popular in the state of California for much of the 1990s and 2000s, the chapter on cells contains a picture of interlocking gears. The caption on the picture is simple: “Cells are microscopic machines.” However, there was nothing in this image or caption to suggest to students that this depiction is anything other than what it is: a metaphor. The simple truth lost to the visions of interlocking cogs is that cells are not, in fact, made of gears.
During those same two decades, in the nationally bestselling Biology textbook by Miller and Levine (which is still used today), the unit on cells is introduced through an elaborate analogy comparing cells to factories. Thirteen components of a cell are given a factory analogue. To give a few examples: DNA is the boss who sits in the central office (nucleus), lysosomes are the janitors removing the “junk” from the factory. The mitochondria are the factory’s power plants. The cell walls are the security perimeter, designed to protect the factory and keep “its products safe and secure” until they are ready to be “shipped out.”
None of this is accidental. Academic research from the early 1990s suggested students learn better when there is an analogue to aid memorizations, and the factory analogy was shown to improve learning when measured through matching tests. Many of these analogies have stuck, and to this day mitochondria are commonly described by science educators or in the popular press as being like power plants or batteries.
To state the obvious: Cells are not factories. Nor do their parts in any way resemble gears. And despite the benefits of using a concept like cells-as-factories and similar analogies as mnemonic learning aids, failing to describe the limitations of those same analogies is likely to result in significant misunderstandings. We know from other research that students just as readily create their own analogical mappings as they use those supplied by the educator—a phenomenon called “over mapping”—and when later asked to recall the lesson, they cannot distinguish between material that was in the supplied analogy and what they independently inferred on their own. In this case, students bring their own varied understandings of factories to supplement the analogy.
If there are thirteen similarities between cells and factories supplied by Miller and Levine, we can easily list many differences.
Machine metaphors didn’t start with high school science curricula. When the 17th century Italian scientist Giovanni Alfonso Borelli dissected animals he described muscles as “inert and dead machines,” furthering the distinction between mind and matter made famous by René Descartes. Meanwhile, Johannes Kepler wished to understand the celestial “clockworks” of the planets. These metaphors were useful in guiding early scientific thought: To the extent that machines could be understood, nature could be as well.
In our everyday language these machine metaphors and analogies have been common ever since then. Whether we need to “blow off steam” after an argument or get a “download” on the party that happened over the weekend, we often use machine metaphors to understand the natural world and ourselves. On reflection, we recognize we are not steam trains that need to blow off the mineral sediments accumulated in the boiler water (the origin of that idiom). Nor can we download information like Neo in The Matrix. That these metaphors remain pervasive may be testimony to our childlike affinity for mechanical engines and how, in some instances such as the picture of gears, the metaphor has become the literal description. Yet, we rarely pause to consider how they may bias our perceptions. Machine metaphors have three potentially negative consequences.
Medicating the machine
One major downside of reducing the human body to a metaphoric machine is that it oversimplifies and gives rise to perceptions that are more mythical than scientific. These simplifications in turn have led to changes in the way health problems may be approached.
Oversimplistic machine models made for an unhelpful dynamic when coupled with willing marketing in the broad expansion of the use of psychotropic drugs over the last 50 years. Because of the pervasiveness of machine metaphors, it was no surprise when Tipper Gore once described her own mental illness as a chemical imbalance where decreased serotonin was a like a car “running out of gas.” On reflection, the notion that brains are like cars and serotonin is like gas is obviously problematic. Even the concept of a chemical imbalance demands our ability to describe what a “chemically balanced” brain looks like—not an easy task.
Fruitful insights may come through challenging culturally assumed ideas of what something is “for” in nature.
This is not to be critical of Tipper Gore, as ultimately any understanding that is helpful to someone managing their own illness should be supported. However, as a general societal understanding of mental illness, the problem with the analogy of the brain and car or an engine is its implied solution. Although antidepressants remain widely used, simply taking an antidepressant to remedy depression in the way you would put gasoline in a car to remedy an empty tank minimizes the profound complexity of mental illness. One of the main classes of antidepressants are selective serotonin reuptake inhibitors—drugs like Prozac and Zoloft. They increase levels of the neurotransmitter serotonin in the brain by preventing its reabsorption, but they don’t do so immediately. Often these drugs take days or weeks to kick in, can sometimes make things worse before they get better, and for many people they never work at all. Filling up your car with gas is an easy fix. Addressing severe depression is not.
Even viewing mental illness as exclusively existing in the brain already omits critical relational, social, and system factors. If mental illness is a consequence of letting the car run out of gas, then could this conception contribute to the stigmatization of mental illness by placing the responsibility exclusively with the individual? Why, after all, wasn’t that person checking their gas gauge and refueling when the fuel light went on?
Likewise, other body-as-machine metaphors are similarly unsatisfactory in their implied scientific explanation. Even our everyday way of speaking about aging or illness often contains a kernel derived from machines. We say “his heart just gave out” like it was the transmission in his car and, in doing so, gloss over the vast differences that exist in longevity among organisms as well as the capacity of living organisms to self-repair.
Designed “for” something
In the Miller and Levine biology textbook described earlier, the unit on evolution contains an illustration of four species of tree finches, each paired with a drawing of a different pair of pliers. The point of the illustration is to show how each finch occupies a distinct niche based on the food it eats and its adaptations for eating that food—illustrated by comparing each beak with a particular type of pliers. The vegetarian finch, for instance, “strips back from woody plants with a beak designed to grip and hold tightly, like a pair of pliers.”
Such descriptions are not uncommon, nor is the use of the word “design” in biology curricula. Just as it is not unusual in a nature documentary for the narrator to describe elephant ears as being for keeping cool on the African savanna or a giraffe’s long neck as being for reaching high leaves.
Yet whenever the words “design” or “for” are used in this way, that neglects the fact that the features of organisms may have many uses in the life of that organism. Birds don’t only use beaks for obtaining food. They also use beaks for grooming, fighting, courtship, feeding their young, singing, heat exchange, and breathing. Selective pressure occurs on the structural feature in relation to the totality of those uses and their composite fitness in a given environment. Moreover, the bird is not a passive recipient of its beak shape but dynamically alters its behavior along the way.
Examples of design thinking can be quite subtle and are not always denoted by the words “design” or “for,” nor do they always explicitly mention the machine analogue that is implied. Fruitful insights may come through challenging culturally assumed ideas of what something is “for” in nature.
Acceptance of natural selection
Perhaps the most profound difference between a factory and a cell is the way the latter came to be. Cellular life evolved through billions of long years and countless generations of creatures facing natural selection pressures. The end products only hint at that long evolution of fits, starts, niche fills, and failures.
Humans can build factories in just a few years or months. And new factories don’t spring from the wreckage of old ones. We clear the land to make them.
The tendency to mechanize biology feeds teleological arguments as well, which can be corrosive to science education. As the theologian William Paley reasoned in 1802, if we find an object that has a complex design, doesn’t this indicate the actions of an intelligent designer?
Yet, to have a complex structure is not equivalent to, and does not require, having a designer. This is not a trivial point. In the United States a nationally representative survey of U.S. high school biology teachers found that 60 percent of them are neither “advocates for evolutionary biology nor explicit endorsers of nonscientific alternatives.” In other words, a majority of the very people we would expect to educate children on natural selection are hesitant to endorse the theory. Science educators and science popularizers unwittingly play into the hands of creationists and their intelligent design arguments against natural selection when they link natural phenomena to machines, factories, and human innovations.
The fundamentally most important aspect of anti-evolutionary creationism today basically comes down to a machine metaphor: Machines are designed by an engineer, and cells are machine-like; therefore cells must be designed by God.
Even if that mapping of biology to manmade machines and therefore to a divine hand is not explicitly stated, it is always implied.
From locomotives to superconductors
The fact that machine metaphors can mislead is not to imply we should ever abandon metaphors and analogies in science education and science discourse altogether. Past efforts to generate a truly objective language stripped of metaphor have been challenged by a diverse group of logicians and philosophers ranging from Kurt Gödel’s pulling the rug out from efforts to establish the logical foundations of mathematics, to Jacques Derrida’s deconstruction of texts that purported to be analytic and objective. Cognitive scientists such as George Lakoff and Mark Johnson meanwhile have shown that many of our abstract terms in language derive from more concrete and tangible concepts. Collectively this work implies that we cannot forgo metaphor and analogy entirely, but we can be more thoughtful in our use of it.
Additionally, while we have discussed machine metaphors and analogies broadly, it is important to remember our great grandmother’s steam engine is not equivalent to our grandchild’s quantum supercomputer. Both steam engines and quantum computers may be labeled “machines” just as both tricycles and supersonic jets may be termed modes of transportation—but these common labels conceal vast differences.
Arguing whether nature is or is not machine-like raises the question of what a machine is. Certainly, at the frontiers of synthetic biology, artificial intelligence, biometric monitoring wearables, and implanted devices that disrupt brains or control glucose, the line between nature and machine is blurring more than at any time in human history. What we can say is that while our old machine metaphors and analogies may have been helpful in furthering science and educating science students, we are long past the point of gleaning insights into the brain by comparing it to an old car engine. The harms of perpetuating these metaphors likely outweigh the benefits.
Mechanical metaphors do not beget wonder towards nature, but instead mistakenly imply we already have the full schematics.
What is needed is to update our metaphors to new and more provocative examples and to correct the myths that stem from past machine metaphors. Returning to the designed factory: In that example the nucleus is the central office and DNA is the “boss.” However, nature has no boss per se. Richard Dawkins introduced the powerful metaphor of a “blind watchmaker” to illustrate how complexity can result in the absence of a plan and a design.
Meanwhile control in the cell is not as hierarchical as the “boss” in the factory metaphor implies. It is more like a well-rehearsed jazz band highly attuned to the signals of the bandmates; playing within general structures and riffing and improving along the way with no conductor directing everything. In the course of a tune, one band member or another may lead and others adapt and adjust just as genetic, epigenetic, or other cellular processes may take the limelight for a time in the life of a cell. Control is decentralized and adaptive—not hierarchical and rigid.
While these metaphors too are imperfect, they challenge what has been the long-standing tyranny of the mechanical metaphor. Furthermore, there are affective dimensions to the mechanical factory metaphors of old because factories are places of rote production and not the dynamic adaptation that may be conjured with a metaphor such as a jazz ensemble. Mechanical metaphors do not beget wonder towards nature, but instead mistakenly imply we already have the full schematics.
Ultimately, we must remember that nature evolved billions of years prior to the machines we would use to understand it. When the metaphor is no longer recognized as such, we risk forgetting that nothing binds nature’s functioning to our machine analogues.
