Some definitions of autonomy
In preparation for the next Life and Mind seminar on the notion of autonomy, I had a look through some of the papers I have at home to see whether I could find any reference to the concept. Since I had a look through them in no particular order, this is reflected in the structure of this post. In any case, this is what I found:
Varela’s ‘Closure Thesis’ states: “Every autonomous system is organizationally closed” (Varela 1979, p. 58). There seem to be two types of closure, namely organizationally and operationally, and both related to autonomy. As Fell and Russell (1994) note: “Autonomous systems are, by definition, operationally closed – or closed by virtue of their organization.” I’m not quite sure how these two types of closure differ, though.
For Varela (1994), “to highlight autonomy means essentially to put at center stage two interlinked propositions: Proposition 1: Organisms are fundamentally the process of constitution of an identity. […] Proposition 2: The organism’s emergent identity gives, logically and mechanistically, the point of reference for a domain of interactions.” This position is repeated in Varela (1997).
“According to the enactive approach, agency and selfhood require that the system be autonomous. An autonomous system is a self-defining or self-determining system, by contrast with a system defined and controlled from the outside or a heteronomous system.” (Thompson 2005). Similarly, Weber and Varela (2002) identify the “‘constitution of an identity’ as the governing of an autonomy principle”.
Quoting from Varela (1979, p. 55-60), Thompson (2005) continues: “An autonomous system is one whose component processes meet two conditions: (i) they recursively depend on each other for their generation and their realization as a system, and (ii) they constitute the system as a unity in whatever domain they exist. An autonomous system can also be defined as a system that has organizational and operational closure: the result of any process within the system is another process within in the system”.
Maturana (1975) claims that “the fundamental feature that characterizes living systems is autonomy” and “that autonomy in living systems is a feature of self-production (autopoiesis)”. Ruiz-Mirazo and Moreno (2000) agree that autonomy in its most basic expression involves not only self-maintenance but also the actual self-construction of a system (e.g. metabolic systems). But what are the material and energetic requirements of such self-production? In a later paper, Ruiz-Mirazo and Moreno (2004) attempt to provide an answer to this question through the “naturalization of the concept of autonomy”.
Boden (1999) states that “metabolism is a type of material self-organization which, […] involves the autonomous use of matter and energy in building, growing, developing, and maintaining the bodily fabric of a living thing.” A similar usage of the word “autonomous” can be found in Moreno and Ruiz-Mirazo’s (1999) paper on metabolism. Her conclusion is that “metabolism is necessary, so strong A-Life is impossible”. I guess it follows from this that ‘strong’ autonomy would be impossible, too. But this still leaves plenty of room for milder versions of autonomy.
As Boden (1996) notes: “Autonomy is not an all-or-nothing property. It has several dimensions, and many gradations. […] An individual’s autonomy is the greater, the more its behavior is directed by self-generated (and idiosyncratic) inner mechanisms, nicely responsive to the specific problem-situation, yet reflexively modifiable by wider concerns.” To unpack exactly what is meant by this would be too much for the purpose of this post, but I recommend everyone having a look at this paper (“Autonomy and Artificiality”).
Wheeler (1997) observes: “The term ‘autonomous’ is commonly used in characterizations of self-organization to capture the fact that the order observed in a self-organizing system has to be explained by appeal to the intrinsic nature of the system itself.”
Barandiaran and Moreno make a further distinction: “The main difference between self-organization alone and the integrity principle of autonomy is that while self-organization appears when the (microscopic) activity of a system generates at least a single (macroscopic) constraint, autonomy implies an open process of self-determination where an increasing number of constraints are self-generated. […] The autonomy of the nervous system is not absolute term but a gradual becoming (unlike Maturana and Varela’s notion of operational closure).”
In Brooks (1991a), we find that one of the principles which defines the domain for their work on behavior-based robotics is stated as follows: “The goal is to study complete integrated intelligent autonomous agents”. From the context it appears that the notion of autonomy is used here in the sense of tether-free robots, where all the energy and computational requirements are stored on board the robots.
In a similar fashion Pfeifer (1996) states as the ‘design principle of autonomous agents’ number 1a: “The agents must be autonomous, i.e. they have to be able to function without human intervention, supervision, or instruction.”
Ok, that’s what I found so far after spending an afternoon going through my papers. I’m sure there’s loads more definitions which I’ve missed. If you know of one which would be worth adding, please leave a comment. I’ll finish with a quotation:
“What science tells us about human autonomy is practically important, because it affects the way in which ordinary people see themselves – which includes the way in which they believe it is possible to behave.” (Boden 1996)