Paul Diduch alerted me to Aristotle’s idea that natural things have an internal principle of motion, while human-made artefacts don’t. That made me wonder how agency and contemporary technology relate to an Aristotelian approach.
In the Physics II.1, Aristotle divides “things that exist” into those that “exist by nature” and those that exist “from other causes” (Hardie & Gaye’s translation):
“[Anything that] exist[s] ‘by nature’ . . . has within itself a principle of motion and of stationariness (in respect of place, or of growth and decrease, or by way of alteration). On the other hand, a bed and a coat and anything else of that sort, qua receiving these designations i.e. in so far as they are products of art—have no innate impulse to change.”
He contrasts things that exist “by nature” with “products of art” or “artificial products”:
“[No] artificial product . . . has in itself the source of its own production. . . . [M]an is born from man, but not bed from bed.”
Artefacts that do have an internal principle of motion
The proposition that artefacts have no innate impulse to change doesn’t seem right. Some non-biological things like smart robots and companies seem to have an internal principle of motion. Thus, there isn’t a clear distinction between something that exists “by nature” and something that is an “artificial product”.
Here are some examples, with implications in square brackets. Phrases in quotes are from Aristotle:
- Scientists create a synthetic bacterial chromosome and introduce it into genomically emptied bacterial host cells that grow and replicate. [The synthetic bacterium is both alive and the result of techne.]
- Polymorphic malware mutates every time it copies itself to a new location to evade detection by anti-virus signature scanners. [The software moves in its software environment and reproduces to avoid threats. Cf. germs that evolve ways to avoid the effects of antimicrobial drugs.]
- You escape from prison and set off over the moors. You’re pursued by guards using a bloodhound, a dog which has been bred since the Middle Ages for tracking people. [The dog’s ability is the result of techne, that is, a breeding program and training.]
- You escape from prison and set off over the moors. You’re pursued by guards using a heat-seeking drone running a self-trained neural net to distinguish an escapee’s tricks and movements from other animals. [The drone’s ability is the result of techne. The techne is self-generated.]
- Company X decides to enter a new market. It develops new products and sales channels. [It has an “innate impulse to change.”]
- The board of Company X proposes to the shareholders that the company should be split. A majority agrees, and companies Y and Z come into being. [A company has been born from a company. It “has in itself the source of its own production.”]
I wonder if Aristotle would still contend, if he were alive today and given these examples, that only entities that exist by nature had an internal principle of motion. Artefacts are more complex now than they were 2,300 years ago. Technology has advanced beyond beds and coats to self-propelled mechanisms that respond to their environment. Even Hero’s machine, a simple radial steam turbine which spins when a water container is heated, wasn’t invented until a few centuries after Aristotle. (Hero’s has “innate impulse to change” if one treats the fire as internal to the thing.)
 |
| Aeolipile, from Knight's American Mechanical Dictionary (1876) via Wikipedia |
Agency vs. “internal principle of motion”
In Defining Agency, I suggested the following criteria for agency, a mashup of various conditions I found in the literature:
An entity that differentiates itself from an environment
… and acts on its own terms,
… responding to the environment
… and adapting to it
Agency and having an internal principle of motion seem very similar, particularly since agents act autonomously, interactively, and adaptively. I would claim that smart robots, software agents, and orgregores (i.e., group agents) have such an internal principle.
Aristotle’s “internal principle of motion” criterion looks like an inference or hypothesis since it couldn’t be tested by observation: the tools for looking inside plants and animals to discern their inner workings, let alone to establish a principle, were lacking in his time. He had to rely on external observation to infer internal workings. Perhaps he was taking an “intentional stance” of predicting behavior by attributing internal states, whether they exist or not (Dennett, 1989). My attempt to recognize agency by only observing behavior rather than making claims about internal states tries to avoid this problem.
Aliveness
Aristotle captures some notion of life or aliveness with his “internal principle of motion and of stationariness” (though he includes the “simple bodies” earth, air, fire, and water in the “by nature” category). Similarly, there is something alive about entities with agency. An internal principle of motion, agency, and aliveness seem like related concepts.
Ginsburg & Jablonka (2019, Table 1) provide a useful inventory of representative twentieth-century lists of characteristics of minimal living systems. They also resonate with Aristotelian concepts, e.g., metabolism in Gánti, and self-production in Maturana & Varela. Non-biological agents and orgregores meet the criteria of at least some of these authors, especially where the characteristics aren’t limited to strictly biological criteria. For example:
Gánti (1971): (1) Inherent unity; (2) metabolism; (3) inherent stability; (4) information-carrying subsystem; (5) program control; (6) growth and multiplication; (7) hereditary system enabling open-ended evolution; (8) mortality
Maturana & Varela (1980): (1) Individuality (closure); (2) self-production; (3) responsiveness; ( 4) regulation and selectivity
The Gánti characteristics, for example, map pretty easily to companies; the trickiest condition might be (7) heredity/evolution, but Articles of Association do change over time. Maturana & Varela’s also seem straightforward to apply, with a suitable definition of what’s included/excluded in a company. These characteristics match well to software agents, too.
Hurricanes
This line of thinking was stimulated by a question of Paul’s about hurricanes (previous post).
I imagine few people today would consider a hurricane to be alive.
I don’t know how Aristotle would classify them. They aren’t artificial products. A hurricane moves of its own accord and so arguably has an internal principle of motion. It even acts to achieve a goal, i.e., minimize imbalances like temperature and pressure gradients. It can spawn more hurricanes (though this is more correct for tornados, I think), so being “the source of its own production.” So presumably a hurricane exists “by nature”?
A hurricane is disqualified from agency by most conventional definitions because it doesn’t have internal representations that encode the state of the world and its goals. However, Floridi & Sanders (2014) consider earthquakes to be agents (“a system, situated within and a part of an environment, which initiates a transformation, produces an effect or exerts power on it”) but not moral agents.
So, do we have here something that isn’t alive, doesn’t have agency, but does have an internal principle of motion?
References
Dennett, D. C. (1989). The Intentional Stance. MIT Press. https://mitpress.mit.edu/9780262540537/the-intentional-stance/
Floridi, L., & Sanders, J. W. (2004). On the Morality of Artificial Agents. Minds and Machines, 14(3), 349–379. https://doi.org/10.1023/B:MIND.0000035461.63578.9d
Gánti, T. (1971, 2003). The Principles of Life (E. Száthmary & J. Griesemer, Eds.). Oxford University Press. https://academic.oup.com/book/4974. Originally published as Az élet princípiuma (1971), Gondolat, Budapest.
Ginsburg, S., & Jablonka, E. (2019). The Evolution of the Sensitive Soul: Learning and the Origins of Consciousness. MIT Press. https://mitpress.mit.edu/9780262039307/the-evolution-of-the-sensitive-soul/
Maturana, H. R., & Varela, F. J. (1980). Autopoiesis and Cognition, in Boston Studies in the Philosophy and History of Science (BSPS, volume 42). https://link.springer.com/book/10.1007/978-94-009-8947-4
No comments:
Post a Comment