Pensée ancienne sur les machines : Différence entre versions

De LERDA
Aller à : navigation, rechercher
Ligne 13 : Ligne 13 :
  
 
The discussion of the production of human-like effects – sounds, for
 
The discussion of the production of human-like effects – sounds, for
example – shows a detailed acquaintance with pneumatic devices.<ref>126 Gregory of Nyssa, de Anima et Resurr. 46.36.36.</ref>
+
example – shows a detailed acquaintance with pneumatic devices.<ref>126 Gregory of Nyssa, de Anima et Resurr. 46.36.36.</ref>  
 
He argues, in essence, that our very ability to formulate this hypothesis undermines the argument.<ref>127 De Anima et Resurr. 2: I thank Stephen Menn for directing me to Gregory. This passage is also discussed in Schiefsky (2007a).</ref>
 
He argues, in essence, that our very ability to formulate this hypothesis undermines the argument.<ref>127 De Anima et Resurr. 2: I thank Stephen Menn for directing me to Gregory. This passage is also discussed in Schiefsky (2007a).</ref>
 
Gregory’s interpretation of the argument focuses on its theological implications rather than its usefulness as a model for natural
 
Gregory’s interpretation of the argument focuses on its theological implications rather than its usefulness as a model for natural
Ligne 90 : Ligne 90 :
 
teleology. He takes the mechanical analogy to show that attributing
 
teleology. He takes the mechanical analogy to show that attributing
 
design to the world need not make the world itself intelligent or
 
design to the world need not make the world itself intelligent or
animate.158 <ref> 158 Lactantius, Div. inst. 2.5.13, cited by Mayr (1986), p. 206 n. 28.
+
animate. <ref> 158 Lactantius, Div. inst. 2.5.13, cited by Mayr (1986), p. 206 n. 28.
 
</ref>He argues that the Stoics are wrong to take the motions
 
</ref>He argues that the Stoics are wrong to take the motions
 
of the heavens as evidence that the stars are living, rather than that
 
of the heavens as evidence that the stars are living, rather than that
Ligne 121 : Ligne 121 :
 
the impulse imparted does not last to infinity but continues after
 
the impulse imparted does not last to infinity but continues after
 
the person giving the motion to the device has stopped, for as long
 
the person giving the motion to the device has stopped, for as long
as the imparted force remains strong.159 <ref>159 Synesius, Aegyptii sive de providentia 1.9.36.  </ref> Synesius' character is focusing on the theology here: he is using the analogy to show how the
+
as the imparted force remains strong. <ref>159 Synesius, Aegyptii sive de providentia 1.9.36.  </ref> Synesius' character is focusing on the theology here: he is using the analogy to show how the
 
gods can infuse a persisting harmony into a world that functions
 
gods can infuse a persisting harmony into a world that functions
 
independently of them.
 
independently of them.
Ligne 135 : Ligne 135 :
 
that 'levering' - to mochleuein - can account for the production of the
 
that 'levering' - to mochleuein - can account for the production of the
 
natural world, because it will not be able to produce the variety
 
natural world, because it will not be able to produce the variety
of shapes and colours found.160<ref>160 Enn. 3.8.2; cf. 5.9.6.  </ref>  In contrast to those who think that the de¯miourgia of nature is like that of wax-modellers, Plotinus  
+
of shapes and colours found. <ref>160 Enn. 3.8.2; cf. 5.9.6.  </ref>  In contrast to those who think that the de¯miourgia of nature is like that of wax-modellers, Plotinus  
 
objects that craftsmen can only make use of existing colours and
 
objects that craftsmen can only make use of existing colours and
 
cannot produce new ones. The techniques of craftsmen are limited
 
cannot produce new ones. The techniques of craftsmen are limited
Ligne 152 : Ligne 152 :
 
anomalous motions of heavenly bodies can be produced from  
 
anomalous motions of heavenly bodies can be produced from  
 
simpler ones with those who try to predict nature using calendrical
 
simpler ones with those who try to predict nature using calendrical
devices.161<ref>161 Proclus, in R. 234.9-22. Sambursky (1962), p. 60.  </ref> A similar complaint about the audacity of trying to provide information about the heavens ahead of time is echoed by Pliny, who regards the heavens as divine.162 <ref> 162 HN 2.9; 2.10. See Bowen (2002a). </ref>Proclus tells us
+
devices. <ref>161 Proclus, in R. 234.9-22. Sambursky (1962), p. 60.  </ref> A similar complaint about the audacity of trying to provide information about the heavens ahead of time is echoed by Pliny, who regards the heavens as divine. <ref> 162 HN 2.9; 2.10. See Bowen (2002a). </ref>Proclus tells us
 
little about the procedures that are used to 'hunt down' the works of nature, or why he objects to this.
 
little about the procedures that are used to 'hunt down' the works of nature, or why he objects to this.
 
   
 
   
Ligne 158 : Ligne 158 :
 
against the work of a mechanic named Theodorus: Proclus' response
 
against the work of a mechanic named Theodorus: Proclus' response
 
indicates that the comparison to a machine was used to argue for
 
indicates that the comparison to a machine was used to argue for
necessitation, apparently in the form of causal determinism.163<ref>163 Proclus, de Providentia; I thank Paul Keyser and Jan Opsomer for the reference. See Borger (1980); Ziegler (2001).  </ref>
+
necessitation, apparently in the form of causal determinism. <ref>163 Proclus, de Providentia; I thank Paul Keyser and Jan Opsomer for the reference. See Borger (1980); Ziegler (2001).  </ref>
 
Theodorus suggested that the world works like a mechanical device,
 
Theodorus suggested that the world works like a mechanical device,
all things depending on the single driving motion.164<ref> 164 Proclus, de Prov. 1.2. </ref> Some language
+
all things depending on the single driving motion. <ref> 164 Proclus, de Prov. 1.2. </ref> Some language
 
in the initial presentation suggests that a comparison is being drawn
 
in the initial presentation suggests that a comparison is being drawn
 
to a theatrical device, although reference is made at the end to some
 
to a theatrical device, although reference is made at the end to some
kind of astronomical calculator.165  <ref>165 parape¯gma, de Prov. 12.65.8. Cf. Steel (2007), pp. 3, 91 n. 279, who notes other references to parape¯gmata in Proclus.  </ref> Proclus tells us little about
+
kind of astronomical calculator. <ref>165 parape¯gma, de Prov. 12.65.8. Cf. Steel (2007), pp. 3, 91 n. 279, who notes other references to parape¯gmata in Proclus.  </ref> Proclus tells us little about
 
Theodorus' account, but evidently the latter saw the potential of
 
Theodorus' account, but evidently the latter saw the potential of
 
mechanical devices to model the idea that complex causal sequences
 
mechanical devices to model the idea that complex causal sequences
Ligne 169 : Ligne 169 :
 
theoretically trained mechanic as well as a philosopher, since Proclus
 
theoretically trained mechanic as well as a philosopher, since Proclus
 
makes a light reference to one of the classic problems of mechanics,
 
makes a light reference to one of the classic problems of mechanics,
'to move a given weight with a given force'.166 <ref>166 De Prov. 4.25.25; Steel (2007), p. 81 n. 117.  </ref>
+
'to move a given weight with a given force'. <ref>166 De Prov. 4.25.25; Steel (2007), p. 81 n. 117.  </ref>
 
Carlos Steel reads the initial references to a theatrical context as
 
Carlos Steel reads the initial references to a theatrical context as
 
metaphorical and suggests that the device used in comparison is a
 
metaphorical and suggests that the device used in comparison is a
clock.167 <ref> 167 Steel (2007), pp. 13-14. </ref> Given the state of technology, however, the initial discus-
+
clock. <ref> 167 Steel (2007), pp. 13-14. </ref> Given the state of technology, however, the initial discus-
 
sion is better taken to refer literally to a theatrical machina, in order
 
sion is better taken to refer literally to a theatrical machina, in order
 
to make the point about deterministic causation by sequential con-
 
to make the point about deterministic causation by sequential con-
 
sequences of a single motion. Theodorus apparently uses the same
 
sequences of a single motion. Theodorus apparently uses the same
 
term that other philosophers use when they point to mechanical
 
term that other philosophers use when they point to mechanical
devices as a model for complex causal sequencing.168<ref>168 eirmon in Latin: de Prov. 1.2.8. See below on the use of automata as models for
+
devices as a model for complex causal sequencing. <ref>168 eirmon in Latin: de Prov. 1.2.8. See below on the use of automata as models for
 
organisms; Steel (2007), pp. 11-12.  </ref> The reference
 
organisms; Steel (2007), pp. 11-12.  </ref> The reference
 
to astronomical devices at the end makes a different point about
 
to astronomical devices at the end makes a different point about
Ligne 183 : Ligne 183 :
 
ambition Proclus criticized in his Republic commentary.
 
ambition Proclus criticized in his Republic commentary.
 
Proclus' criticism of Theodorus' programme includes a reference
 
Proclus' criticism of Theodorus' programme includes a reference
to the presence of 'powers' in organisms.169 <ref> 169 De Prov. 3.11. See Steel (2007), p. 76 n. 48. </ref> The argument here is
+
to the presence of 'powers' in organisms. <ref> 169 De Prov. 3.11. See Steel (2007), p. 76 n. 48. </ref> The argument here is
 
compressed, but it is perhaps invoking a Galenic reading of the
 
compressed, but it is perhaps invoking a Galenic reading of the
 
Aristotelian idea that organic nature requires internal powers of
 
Aristotelian idea that organic nature requires internal powers of
 
self-maintenance. The idea may be that these specific internal powers
 
self-maintenance. The idea may be that these specific internal powers
cannot be accounted for by structural rearrangements of the parts,170 <ref>170 See Galen's criticism of the Erasistratean materialists, above  </ref>
+
cannot be accounted for by structural rearrangements of the parts, <ref>170 See Galen's criticism of the Erasistratean materialists, above  </ref>
 
but rather - as Plotinus argued against the proponents of 'leverage' -
 
but rather - as Plotinus argued against the proponents of 'leverage' -
 
must be produced by qualitative alterations, giving rise to irredu-
 
must be produced by qualitative alterations, giving rise to irredu-
Ligne 197 : Ligne 197 :
 
might have been that the mechanical view of causal sequencing as
 
might have been that the mechanical view of causal sequencing as
 
material interconnections driven by a single mechanism misses the
 
material interconnections driven by a single mechanism misses the
explanatory autonomy that must be accorded to organisms.171<ref>171 De Prov. 3.11: I take this to be the point of the reference to Aristotle's claim that
+
explanatory autonomy that must be accorded to organisms. <ref>171 De Prov. 3.11: I take this to be the point of the reference to Aristotle's claim that
 
organic processes can go against fate. Cf. Steel (2007), p. 76 n. 50. </ref> A
 
organic processes can go against fate. Cf. Steel (2007), p. 76 n. 50. </ref> A
 
machine model of the universe would tend to erase the explanatory  
 
machine model of the universe would tend to erase the explanatory  
Ligne 206 : Ligne 206 :
 
calculators shows the existence of incorporeal ideas in the mind of
 
calculators shows the existence of incorporeal ideas in the mind of
 
the craftsman, thus - presumably - showing that not every kind of
 
the craftsman, thus - presumably - showing that not every kind of
cause can be accounted for by the machine analogy.172<ref>172 De Prov. 12.65. </ref> The argument
+
cause can be accounted for by the machine analogy. <ref>172 De Prov. 12.65. </ref> The argument
 
seems to be that a machine is material, so - if machines cannot
 
seems to be that a machine is material, so - if machines cannot
 
include incorporeal ideas - no machine could construct a machine.
 
include incorporeal ideas - no machine could construct a machine.

Version du 12 juillet 2017 à 20:39


SYLVIA BERRYMAN :

"THE MECHANICAL HYPOTHESIS IN ANCIENT GREEK NATURAL PHILOSOPHY"


Christian authors from the same period seem to have taken a particular interest in the machine analogy. One of the most interesting versions of this is that of Gregory of Nyssa (fourth century CE), who proposes a mechanical analogy and presents the criticisms of it by his teacher Macrina. Gregory is interested in the evidence that the soul is immortal and has a genuine, immaterial rationality. He takes the point of the machine analogy to be that of suggesting how human beings might work without rationality, by purely material means.

The discussion of the production of human-like effects – sounds, for example – shows a detailed acquaintance with pneumatic devices.[1] He argues, in essence, that our very ability to formulate this hypothesis undermines the argument.[2] Gregory’s interpretation of the argument focuses on its theological implications rather than its usefulness as a model for natural philosophy.[3]He does not, ultimately, reject the claim that our devices imitate the functions of organisms, perhaps including ration?ality, but rather points to the capacity to create them as evidence of human reason:

"We see many such things contrived by the makers of machines, in which they arrange matter skillfully to imitate nature. Their contrivances do not show similarity to nature in appearance alone, but also sometimes in motion, and in representation of a kind of voice, when the mechanism reverberates in its sounding part. In such cases indeed the phenomena do not lead us to suppose that an intelligent power brings about in each machine the appearance, form, sound, or motion. If we should say that the same also happens in the case of this mechanical instrument of our nature, we might say that no intelligent essence is infused in us according to the peculiarity of our nature, but some kinetic power resides in the nature of the elements in us. Such activity would be a result.[4]"

Here, a mechanical analogy is considered full-blown: while it is primarily directed to the functioning of an organism, it is clear that it could apply a fortiori to nature as a whole. Not only organic functions but also intelligent behaviour can be imitated by an unthinking mechanism. Devices had been built that do something like the kinds of things that organisms do; and it was at least a conceptual possibility that we are just like them.

The issue that Gregory is concerned about is the existence of an immortal and separable soul, conceived as rationality. The machine analogy serves to raise the possibility that we might be composed entirely from unintelligent material nature, merely programmed to look as though we are acting rationally. Macrina’s response to this challenge is not to offer evidence that any given human being is rational – to solve the problem of other minds – but rather to suggest that the human ability to create such devices – devices that imitate purposive behaviour – itself demonstrates the existence of human intelligence, in at least some humans. We first study processes in nature to learn the capacities of materials and then conceive devices that exploit these capacities. The very fact that we can build devices that imitate the functioning of organisms shows not that we might work like machines, but that we have access to an immaterial rationality. Given the evident assumption that rationality is immaterial, the point seems to be that an immaterial cause is needed to account for the causal action of at least some human beings, those capable of craft activity. Thus – presumably – since machines are arrangements of matter, the human craftsman is not a machine.[5]

Macrina suggests that the materialist position amounts to a claim that the matter could organize itself into working artifacts.[6] The implication seems to be that the attempt to deny the need for an additional immaterial soul to account for human capacities – by build?ing the capacity for goal-directed activity into matter itself – should entail matter’s being able to produce functioning artifacts automati?cally. Those who erase the distinction between natural and artificial, that is, need to explain why artifacts are not self-organizing. As both Epicurus and the Stoics are mentioned as philosophical targets, this counter is presumably a critique of the Stoic idea that rationality is spread throughout all matter as well as a critique of non-teleological materialism. In contrast to the Cartesian move of looking inward to establish our awareness of the activity of mind, Gregory seems to be relying on the external evidence of rationality in our ability to engineer effects. Gregory turns the ‘machine analogy’ on its head and finds proof of human intelligence in the building of the device.

_____


  1. 126 Gregory of Nyssa, de Anima et Resurr. 46.36.36.
  2. 127 De Anima et Resurr. 2: I thank Stephen Menn for directing me to Gregory. This passage is also discussed in Schiefsky (2007a).
  3. 128 On the fusion of Platonic, Aristotelian and Stoic elements in Gregory’s philosophical background, see Zachhuber (2000), p. 151.
  4. 129 Gregory of Nyssa, e Anima et Resurr. 46.33.11. Translation by Roth (1993), pp. 40–1.
  5. 130 Compare Proclus’ argument against Theodorus, above.
  6. 131 Gregory of Nyssa, de Anima et Resurr. 46.37.33.


157 For example, Seneca (first century CE) uses analogies to technology in an unsyste- matic way, and alongside organic analogies. A siphon analogy is embedded in the comparison to organisms (Q Nat. 2.6; 6.14); he uses an organic analogy to explain why water emerges continuously from springs and why rivers flow (Q Nat. 3.15). He doubts that the capabilities of artifacts merit comparison: natural forces are better able to create an upward flow than any artificial method (Q Nat. 2.9)

____


The Stoic appeal to a cosmic machine analogy may make most sense if it is regarded as an ad hominem response to others appealing to the machine analogy to undercut the argument for divine design. The Stoics would not be initiating the use of the machine analogy, but turning it against their opponents. At least one Christian critic responded in kind. Lactantius (third century CE), draws on Archimedes' sphere in order to show the errors of Stoic immanent teleology. He takes the mechanical analogy to show that attributing design to the world need not make the world itself intelligent or animate. [1]He argues that the Stoics are wrong to take the motions of the heavens as evidence that the stars are living, rather than that their motions are imparted to them by God's design. The comparison to Archimedes' device here is implicitly used to support the possibility that a sequence of regular and orderly motions could have been produced in a non-living thing by design. He needs to show that there is an alternative to supposing that the motions of the heavens are either actively guided by intelligence on an ongoing basis or happen by chance. Here the possibility of a machine designed to operate without intelligence, yet in a fashion that exhibits design, is used to argue for a particular view of God's relationship to the cosmos. Although Lactantius does not go on to exploit the implications of this model for the study of the natural world, he clearly draws the analogy between cosmic simulacra and the heavens in order to further the idea that the complex and interconnected motions of the heavens could be constructed to exhibit order without ongoing intelligent direction.

Another Christian writer, Synesius of Cyrene (fifth century CE), a student of Hypatia, takes up a similar, if more modest, use of the machine analogy. Rather than think of the cosmos as a whole as a machine, Synesius appears to use the analogy as a way to conceive of divine intervention in the cosmos: he puts into the mouth of an Egyptian sage the idea that divine intervention occurs on an occa- sional basis, in the same way as power is imparted to theatrical devices, ta neurospasta organa. Although he describes the devices as tools, organa, it is clear from the context that he is describing working artifacts of considerable complexity. The claim is that the impulse imparted does not last to infinity but continues after the person giving the motion to the device has stopped, for as long as the imparted force remains strong. [2] Synesius' character is focusing on the theology here: he is using the analogy to show how the gods can infuse a persisting harmony into a world that functions independently of them.

Although these two Christian sources have some sympathy with this appeal to mechanics as a way to conceive of divine action, it met with criticism in the Platonic and Aristotelian traditions. The idea of a cosmic craftsman has a natural home in the Platonic tradition, and there is some evidence that Platonists of late antiquity resisted the attempt to interpret Plato's de¯miourgos as too much like an engineer. Plotinus (third century CE) attacks the idea that construction techni- ques might suffice to explain the natural world. He rejects the idea that 'levering' - to mochleuein - can account for the production of the natural world, because it will not be able to produce the variety of shapes and colours found. [3] In contrast to those who think that the de¯miourgia of nature is like that of wax-modellers, Plotinus objects that craftsmen can only make use of existing colours and cannot produce new ones. The techniques of craftsmen are limited to reshaping and structuring material: they cannot turn straw into gold. The Neoplatonist tradition accepted Aristotelian arguments against the irreducibility of qualitative change, and, like Galen, was concerned to preserve the notion of qualitative transforma- tion in order to account for the functions of organic natures in particular.

Proclus (fifth century CE) also complains about investigators' use of devices to try to understand the processes of the natural world; he mentions alchemists as well as astronomers. In a passage noted by Sambursky for its defence of holism in nature, Proclus lumps together the mathematicians who try to show how the apparently anomalous motions of heavenly bodies can be produced from simpler ones with those who try to predict nature using calendrical devices. [4] A similar complaint about the audacity of trying to provide information about the heavens ahead of time is echoed by Pliny, who regards the heavens as divine. [5]Proclus tells us little about the procedures that are used to 'hunt down' the works of nature, or why he objects to this.

Proclus directs an entire treatise - preserved only in Latin - against the work of a mechanic named Theodorus: Proclus' response indicates that the comparison to a machine was used to argue for necessitation, apparently in the form of causal determinism. [6] Theodorus suggested that the world works like a mechanical device, all things depending on the single driving motion. [7] Some language in the initial presentation suggests that a comparison is being drawn to a theatrical device, although reference is made at the end to some kind of astronomical calculator. [8] Proclus tells us little about Theodorus' account, but evidently the latter saw the potential of mechanical devices to model the idea that complex causal sequences follow automatically from a single cause. Theodorus seems to be a theoretically trained mechanic as well as a philosopher, since Proclus makes a light reference to one of the classic problems of mechanics, 'to move a given weight with a given force'. [9] Carlos Steel reads the initial references to a theatrical context as metaphorical and suggests that the device used in comparison is a clock. [10] Given the state of technology, however, the initial discus- sion is better taken to refer literally to a theatrical machina, in order to make the point about deterministic causation by sequential con- sequences of a single motion. Theodorus apparently uses the same term that other philosophers use when they point to mechanical devices as a model for complex causal sequencing. [11] The reference to astronomical devices at the end makes a different point about our capacity for foreknowledge based on calculation. This is the ambition Proclus criticized in his Republic commentary. Proclus' criticism of Theodorus' programme includes a reference to the presence of 'powers' in organisms. [12] The argument here is compressed, but it is perhaps invoking a Galenic reading of the Aristotelian idea that organic nature requires internal powers of self-maintenance. The idea may be that these specific internal powers cannot be accounted for by structural rearrangements of the parts, [13] but rather - as Plotinus argued against the proponents of 'leverage' - must be produced by qualitative alterations, giving rise to irredu- cibly teleological powers specific to the organism. The Aristotelian view of the organism as a self-maintaining unit is defended: the irreducibly teleological role of natures gives organisms a special kind of unity and cohesion that cannot be accounted for by thinking of them as parts of a greater interconnected whole. Proclus' concern might have been that the mechanical view of causal sequencing as material interconnections driven by a single mechanism misses the explanatory autonomy that must be accorded to organisms. [14] A machine model of the universe would tend to erase the explanatory boundary between organism and environment, since it accords no special causal priority to the natures of substantial individuals. Proclus suggests a different critique of the machine analogy at the end of the treatise. He argues that the possibility of constructing calculators shows the existence of incorporeal ideas in the mind of the craftsman, thus - presumably - showing that not every kind of cause can be accounted for by the machine analogy. [15] The argument seems to be that a machine is material, so - if machines cannot include incorporeal ideas - no machine could construct a machine.

As in Gregory, the very nature of design is used to show that we, as artisans guided by incorporeal ideas, could not be machines. The machine analogy seemed to have some play in the cosmo- logical debates of late antiquity, both as a way to think about divine intervention as an occasional rather than ongoing causal interven- tion, and as a model for complex causal sequencing. Although the philosophers recording these analogies were often those rejecting them, there was clearly some interest in this way of viewing causal transmissions. Mechanical devices do seem to be exploited to make a more general point about the relationship between complex causal sequences and ongoing intelligent direction, a point equally appli- cable to the case of the heavens or to aspects of the natural world.

Just as the author of de Mundo appealed to a machine in order to suggest that a god designed the world to run by itself, Lactantius rejects the idea that the regularities of the heavens entail the presence of intelligence internal to the bodies exhibiting those regularities.

The machine analogy is used to support the idea that compl ordered sequences could be brought about - in designed complexes - by material interactions and thus speaks against those who suppose that teleology requires us to posit ongoing intelligent direction. The same issue of causal sequencing can be seen in the texts comparing working artifacts to organisms.

Along with the idea that the cosmos as a whole is a kind of device constructed to run without ongoing intelligent direction comes the possibility that we ourselves are part of that matrix. Gregory and Proclus explicitly respond to this possibility; Marcus Aurelius' reference to non-sages as puppets implicitly depends on a similar point. Thus, Christian, Stoic and Platonist alike contrast their notion of the rational capabilities of human beings to the mechan- ical hypothesis. The rejection of the machine analogy had a partic- ular importance for those concerned to mark out the possibility of a specifically human capacity to transcend the capacities of the natural world and to stand in a particular relationship with the divine.

These responses show the limits to the potential of the machine analogy as a guiding heuristic in investigating organic nature. I hope to have demonstrated that, against the background of Hellenistic mechanics, the machine analogy could clearly be formulated. Although we know more about the philosophical position of its detractors, it evidently existed as a hypothesis in late antique natural philosophy, distinct from both the anti-teleological materialism of the atomists and the irreducibly teleological powers of a Galen or a Neoplatonist. While this hypothesis does not seem to have had a large following, it evidently attracted some attention. The specific reasons why it was rejected by the philosophical schools deserve attention, since they are evidently more interesting and more specific than a separation in kind between art and nature.

Ancient Greek mechanics offered working artifacts complex enough to suggest that the natural world might work in similar ways. The rejection of the mechanical hypothesis can tell us much about the interpretation of teleology in late antiquity, about the conception of causal sequences, and about the conception of the relationship between matter and form. Mechanics built devices that worked, raising new possibilities about what results could be achieved by structural arrangements of matter. Ancient Greek nat- ural philosophers did not simply ignore these attempts at devising nature. There were those in late antiquity who considered a mechan- ical hypothesis and wondered whether organisms, the cosmos as a whole, or we ourselves, might 'work like that'.

_____


  1. 158 Lactantius, Div. inst. 2.5.13, cited by Mayr (1986), p. 206 n. 28.
  2. 159 Synesius, Aegyptii sive de providentia 1.9.36.
  3. 160 Enn. 3.8.2; cf. 5.9.6.
  4. 161 Proclus, in R. 234.9-22. Sambursky (1962), p. 60.
  5. 162 HN 2.9; 2.10. See Bowen (2002a).
  6. 163 Proclus, de Providentia; I thank Paul Keyser and Jan Opsomer for the reference. See Borger (1980); Ziegler (2001).
  7. 164 Proclus, de Prov. 1.2.
  8. 165 parape¯gma, de Prov. 12.65.8. Cf. Steel (2007), pp. 3, 91 n. 279, who notes other references to parape¯gmata in Proclus.
  9. 166 De Prov. 4.25.25; Steel (2007), p. 81 n. 117.
  10. 167 Steel (2007), pp. 13-14.
  11. 168 eirmon in Latin: de Prov. 1.2.8. See below on the use of automata as models for organisms; Steel (2007), pp. 11-12.
  12. 169 De Prov. 3.11. See Steel (2007), p. 76 n. 48.
  13. 170 See Galen's criticism of the Erasistratean materialists, above
  14. 171 De Prov. 3.11: I take this to be the point of the reference to Aristotle's claim that organic processes can go against fate. Cf. Steel (2007), p. 76 n. 50.
  15. 172 De Prov. 12.65.