Teleology in Biology: A Value-Based Approach.

Teleology in Biology

A Value-Based Approach

MA-Thesis Victor Berndsen Dr. H. van den Berg University of Amsterdam (UvA)

1. Teleology in Biology: How & Why

1.1 General introduction

Imagine a pasture occupied by a small herd of spotted cattle. The cows are dispersed over the field, peacefully ruminating and going about their business. Suddenly, the sky darkens and thunderclouds start to form over the meadow. After some time lightning strikes in the field, causing one of the bovine animals to perish. However, lightning is not the only peril the herd has to face this day. The blackening of the sky has prompted a hungry carnivore to seize the opportunity. Amidst the chaos of rain and thunder, a lone wolf ambushes an unfortunate grazer. This results in yet another casualty amongst our imaginary herd. Next morning, the herdsman arrives to check upon his cattle. To his surprise he finds two carcasses, one stripped of its tender meat and the other completely charred. The question he now asks himself is: how did these cows die?

Given the knowledge that we have, the answer is simple: the charred cow died because it was struck by lightning, the other because it got preyed on by a wolf. If we desire to know more about the cause of death, that is, to dive deeper into the explanation of the causal factors at play in the demise of these two cows, we would have to appeal to science. For example, in the case of the lightning strike, we could consult a meteorologist. The

meteorologist will be able to explain in detail the mechanism of lightning. He or she will refer to the condensation of water vapor which results in the formation of cloud. Here, electrically charged regions come into being. The electric potential between these regions will

eventually cause a discharge in the form of lightning, which carries a more than lethal charge (hopefully, the meteorologist’s account will be a little more comprehensive than mine). For additional information about the other case, we will have to turn to a biologist’s expertise. Similar to the meteorologist, the biologist is able to give a comprehensive account of the inner workings of a lupine animal. For example, he or she could refer to the physical

structure of the wolf, the workings of its organs, its characteristic features and its behaviour. However, I am inclined to think that an exposition of the wolf’s bloodstream, digestive system or energy maintenance will leave one quite dissatisfied when it comes to explaining the demise of our unfortunate cow. For the explanation provided thus far leaves one intuitively important question unanswered: why did the wolf kill the cow?

What I aim to show with this (rather improbable) scenario is the particular type of questioning that seems appropriate for a biologist, while less so for a meteorologist or a physicist. After all, it makes no sense to ask why the cow was struck by lightning. Non-living natural

however cannot be explained exhaustively without referring to some form of a ‘why’ question. Evolutionary biologist and philosopher of science Ernst Mayr explicated this distinction between ‘how’ and ‘why’-questions in an article that remains influential up until today (Mayr 1961, Laland et al. 2011, 2013). He defines ‘how’ questions as referring to proximate causes, whereas ‘why’ questions employ ultimate causes1. Proximate causes are studied in all fields of natural science: physics, chemistry and biology alike. They are

proximate (direct) in the sense that they explain isolated, functional mechanisms. Proximate causes are studied in the part of biology that Mayr calls ‘functional biology’, where organs, cells, or organelles are isolated and analyzed in terms of the physical and chemical

processes that constitute them. An example from physics would be the (oversimplified) account of lightning that I have provided. Organisms can be analyzed in terms of proximate causes as well, that is in terms of the biochemical processes and physical structures they are composed of, but in the case of the lone wolf we have noted that this leaves one quite dissatisfied. Mayr explains why, for ‘every organism (...) is the product of (...) a history which indeed dates back more than 2000 million years’. He continues: ‘there is hardly any structure or function in an organism that can be fully understood unless it is studied against this

historical background’ (Mayr 1961, 1502). Unlike in chemistry or physics, there are no ‘absolute phenomena’ in biology, as ‘everything is time-bound and space-bound’ (ibid.). This leads Mayr to distinguish proximate causes from ultimate (distant) ones. Ultimate causes employ a historic chain of events, unlike proximate causes. Moreover, ultimate causes are unique to biology (specifically to evolutionary biology), while proximate ones are not. In our example, the description of thunder is a proximate cause. The physical composition of the wolf is a proximate cause as well. However, no explanation of the wolf’s behaviour is complete without referring to (at least in part) to the history of the wolf’s coming-to-be. The wolf’s behaviour is dynamic in its local and historical contingency, while the behaviour of thunder is lawlike and static (no pun intended). This is the reason that ‘why’-questions and ultimate causes (which are employed in the answers to them) are unique to living things.

Mayr’s proximate/ultimate distinction serves to reinforce the idea that biology as a science is unique and differs from other natural sciences in a fundamental manner. Where physicists study fixed and isolated phenomena, the objects of biological enquiry, organisms mainly, are historic and local contingencies. Lightning strikes emerge all around the globe (and maybe even on other planets, given that comparable physicochemical conditions apply) on the

1 _{As an aside, it is good to note that Mayr uses the word ‘cause’ in a different sense than what we are} familiar with in common speech. Instead of ‘cause’ as a ‘person, thing, or event mechanically making something happen’, Mayr can be taken to employ it in the sense of a ‘basic explanatory factor’, or simply ‘explanation’ (Mix 2016, 1818).

basis of the same mechanism, and have always done so. Now, in contrast, think of the textbook version of what prompted the idea of evolution through natural selection in the mind of Charles Darwin. The birth of his theory is often presented through his discovery of

different birds from the same species, on different neighbouring islands. He found that these birds displayed minute but observable differences in the shapes and sizes of their beaks. What Mayr means by ultimate causation is that these differences can only be explained by taking into account the historical and regional conditions in which this avian species evolved. For example, the accessibility of worms is of obvious great importance to birds. Long, narrow beaks are useful to dig up worms out of the sand. However, on an island that has a more rocky terrain, this type of beak is not as useful, as worms are not accessible through a rocky surface. On this type of terrain a smaller, broader and stronger beak is more useful to the bird as other types of nourishment are more easily available, like nuts or shellfish. When we ask why there exist these differences between the beaks of birds from the same species, proximate causes do not suffice as an explanation, and ultimate causes need to be

employed as well in order to fully understand the phenomenon at hand (Mayr 1961, 1503).

Within ‘why’-questions, which are so characteristic of biology as they invoke ultimate causes, Mayr detects a certain ambiguity. ‘Why?’ may mean ‘how come?’ but it can also mean ‘what for?’. The first type of ‘why’-question being historical, and the second type being finalistic in the sense of referring to a goal or purpose (Mayr 1961, 1502). ‘What for’-questions employ explanations that invoke ends, goals or purposes. ‘Why am I writing this thesis? In order to obtain a master’s degree in philosophy.’ I have a certain goal in mind, and my behaviour can only be explained in light of this end. Me staring at a computer screen and reading through countless books and journals for hours at a time is clearly purposeful behaviour. So, I would argue, is the behaviour of the wolf when it is stalking a cow. I would even go as far as assigning purposeful behaviour to organisms that cannot be said to maintain intentions, for example because they do not possess anything resembling a brain. A virus, for example, infects its host by injecting genetic material into the nucleus of a cell for the purpose of reproduction. This way of talking about organisms (leaving aside the discussion on whether a virus can be labelled an organism at all, given that it consists of little less than a few protein molecules) makes perfect sense to me and seems completely appropriate. Not only behaviour, but other biological features like growth and reproduction can labelled purposive, or teleological as it is classically referred to. However, Mayr only wants to go as far as labelling these processes as seemingly purposive (Mayr 1961, 1503). He deems it obvious that the biologist only has (and should have) the ‘how come’-question in mind. This implies that ‘what for’-questions are dubious from a scientific standpoint. He thinks of apparent purpose as opposed to scientific thought (Mix 2016, 1827-28). I take this intuition to be

shared by many, and I hope the reader is one of them. And, although I do not agree with it, I reckon it is an understandable standpoint. For I take it to be obvious that talking about purposes, goals and ends as part of the natural order, sounds suspicious at the least. God had bestowed nature with purpose. And before that, Ancient Greek philosophers had proposed immaterial forces to ground end-directedness (teleology) in nature. We will delve into these traditions later. For now, it suffices to note that it is common sense that modern science should not rely on such notions of intelligent designers or immaterial forces. Thus, received opinion seems to be that there is no room for these kind of explanations in

contemporary biology (e.g. Walsh 2008, 2). For example, David Hull, one of the pioneers of philosophy of biology, concludes that: “from the point of view of contemporary biology, both vitalism and teleology are stone-cold dead” (Hull 1969, 249). Paradoxically, biological language, both in common speech and in academic discourse, is brimming with seemingly teleological notions (e.g. Walsh 2008, Trestman 2012, Kampourakis & Minelli 2014). Modern biologists tend to shy away from using words like ‘goal’, ‘purpose’ or ‘end’. Instead, they often employ more modern terms like ‘function’ or ‘design’, which are just as teleological (Allen 2009). And it is natural language for laymen as well, as I do not think that anyone would object to the statement ‘the wolf stalked the cow for the purpose of killing it’. Given this paradox, it remains a controversial subject of philosophical enquiry as to what role there is left to play for teleological explanations in biology (e.g. Mix 2016, 1818, Wattles 2006, 446, Zammito 2006, 749). Can teleology be a legitimate part of scientific explanation? This will be the central question of this thesis, and by now it should come as no surprise that I will

attempt to build a case for a positive answer.

1.2 Methodology and summary of content

I have introduced two notions that I have hitherto left undefined, being mechanism and teleology. Before we move on, we have to demarcate what is meant exactly by these terms. The concept of mechanism has seen a lot of different uses in different contexts, and it can be deployed in a vast number of different ways (Craver & Tabery 2017). It is not my intention to delve into the complex debates surrounding it, as this would exceed the aim of this thesis. For us, it suffices to note that since Descartes, modern natural sciences have been

dominated by a universal explanatory model that has been dubbed mechanistic (Walsh 2008, 3). In philosophy, the term new mechanism has been coined for the emergence of a framework for thinking about the philosophical assumptions underlying many areas of science. Emphasis on the importance of mechanism has been historically associated with the rejection of teleology (Cravery & Tabery 2017). Therefore, it is quite common to present mechanism and teleology as mutually exclusive, opposing modes of explanation, and mechanistic explanations are clearly preferred over teleological ones by standards of

modern science. Throughout this thesis, we will work from this opposition. At the one hand we have mechanism, which we take to be the modern prescribed model for natural science. These explanations apply to everything in nature and refer to physical structures and

chemical properties. In Mayr’s terms, these explanations employ proximate causes only. Later on, we will see that within Aristotle’s framework, mechanistic explanations are limited to efficient causes. In short, mechanistic explanations employ causal mechanisms. A phenomenon is explained by analyzing causal relations between different parts of a

phenomenon. The prime example of a mechanism is a clockwork: one is able to explain it by identifying the different cogs and wheels and then analyzing the causal relations between them (one cog moves the other and so forth). On the other hand we have explanations that refer to goals, ends and purposes, ‘for-the-sake-of-what’-kind of explanations. We will name this type of explanation teleological. ‘Teleology’ is composed out of two Ancient Greek roots: telos and logos. The root ‘logos’ is quite well known and is often translated with ‘word’, but in common usage it is taken more broadly as referring to anything that has to do with

‘explanation’ or ‘study’. For example, ‘psychology’ refers to the study of the psyche (mind), and ‘biology’ attempts to explain bios (life). ‘Telos’ is often translated as ‘end’, ‘goal’, or ‘purpose’. Therefore, teleology is the study of ends, or of end-directed explanations. The term ‘teleology’ or ‘teleological’ will be used quite extensively throughout this thesis, and I will always employ it in reference to a mode of explanation that deploys goals, end or purposes. In short, the dichotomy that underlies this thesis is as follows: on the one hand we have mechanism, where phenomena are explained by referring to underlying mechanisms and proximate causes. On the other we have teleology, where phenomena are explained in reference to goals, ends, purposes and so forth. While mechanical explanations in science are quite unproblematic, teleological explanations have a more dubious status as to whether they can and should be a legitimate part of the scientist’s explanatory toolkit. I will argue that this dubious status is unjust and unreasonable.

In chapter two, we will explore the historical dimension of the problematic relationship between teleology and mechanism. We will do so in line with Michael Ruse (2017). He distinguishes three modes of teleology, which can be linked to three major figures in the history of philosophy. There is external teleology, which originated with Plato. Secondly, Ruse identifies internal teleology, as maintained by Aristotle. Thirdly there is heuristic teleology, which Ruse traces back to the Ancient Greek atomists, but comes to full bloom in the work of Immanuel Kant. The goal of this chapter is twofold. At first, we will trace

teleological thinking (and the modern aversion against it) back to its origins. We will see that, especially in the case of external teleology, explaining things in terms of their goals or

historical connection renders teleology suspicious in modern times from a scientific

viewpoint. Secondly, as a contrast, I will show that the tradition of heuristic teleology is not inherently opposed to modern science and can be reconciled with it. As I hope to show, the door is not entirely closed for teleology in biology.

In chapter three, we will explore a structural problem with teleology that deems it unfit to be reconciled with modern science. This is the problem of backward causation. While ‘regular’ mechanistic scientific reasoning presupposes causes to precede their effects in time, teleological explanations are often taken to have effects precede causes. For a goal or a purpose is almost always something that lies in the future, and may not yet be accomplished while its effect is taken for granted. Teleological explanations are therefore taken to refer to future events. Through the works of (mainly) Francisco F. Ayala and Ernst Mayr, we will see if this problem can be solved in order to take another hurdle in our project of providing teleology with scientific respectability. For backward causation stands in opposition to mechanistic causality where causes precede their effect. If we can build a case for an interpretation of teleology that does not rely on backward causation, it will be more acceptable given the standards of modern science.

In the fourth and final chapter of this thesis we will explore a relatively new and original approach to teleology in biology. We will do so through the works of Mark Bedau and Michael Ruse. This approach is based in value, and works from the premise that every organism has a good of its own. While unscientific at first sight, a value-based approach to teleology in biology might open up possibilities to reconcile it with modern scientific

2. External, Internal and Heuristic Teleology

2.1 General introduction

Most histories of philosophy start with Plato, and so will this preliminary ‘history of teleology’. With Plato, we will investigate the mode of teleology that is commonly referred to as

external, extrinsic or transcendent. I will show why I believe this interpretation proves most problematic to the modern reader, and why the unsettling connotations of teleology can be traced back to Plato. As is the case in most philosophy textbooks, from Plato we will move on to Aristotle. Here we will explore a mode of teleology that is often introduced as internal or immanent. We will see that this way of explaining purpose in nature causes less

discomfort in the modern reader than does Plato’s. However, internal teleology still causes some discomfort, which is where our third central figure comes in: Immanuel Kant. His take on teleology in nature will be expounded, which we will name (with Ruse) heuristic teleology. Finally we will conclude that his analysis of teleology is quite acceptable within the

framework of modern mechanistic science, and therefore most promising for the aims of our project.

2.2 External teleology: Plato

As is well known, Plato wrote on a wide range of topics in around thirty works that are known as dialogues. The dialogue is a particular style of prose that is very characteristic of Plato, wherein two or more characters exchange their views on a specific topic. The most well-known dialogue is without much doubt the Republic, where he expounds his famous allegory of the cave, as well as his views on ethics and politics. Of much more interest to us is the Timaeus, where Plato sets forth his cosmology and his analysis of purpose in nature. He does so in the person of the eponymous leading figure of the dialogue, whose doctrines we can simply regard as Plato’s own (Plato 1937, 3). He does not only explain the ordering of the cosmos, but also the creation of it. As with the cosmogony of Genesis, debate surrounds Plato’s creation myth on whether it is to be taken literal or as a metaphor(Zeyl & Settler 2017). Interpretational discussions aside, ‘in a nutshell, it [the Timaeus] is a story of a

Creator who ventures to replicate, in time and space, an image of eternal, heavenly patterns’ (Wattles 2006, 447). Note that, in contrast to the Creator-from-nothing in the Judeo-Christian tradition, the Platonic Creator is a designer from the already-existing (Ruse 2017, 1:22). First there was chaos, which is then shaped by the Creator towards order.

Plato names his Creator demiurge, which is often translated as Craftsman. The eternal immaterial pattern that the demiurge ventures to replicate is nous, which is often translated with Mind. For the modern reader, the concept of nous is difficult to grasp. Moreover, the

relation between nous and the demiurge is highly complex and object of academic

disagreement (Zeyl & Settler 2017). An analysis of this relation greatly exceeds the scope of this thesis. It will suffice for us to note that nous is an ordering principle, an immaterial force that drives the unfolding of the universe from chaos towards goodness or perfection, with the demiurge as its executor. In other words, at least in my view, the demiurge streamlines and regulates the ordering principle that is nous. The demiurge is an active agent that embodies this principle, which shapes the cosmos in an ongoing transition from chaos to perfection. The demiurge is an intentional agent as he has an obvious goal ‘in mind’. He streamlines the unfolding of the universe with the purpose of emulating the good and perfect divine plan that is nous. The demiurge strives towards maximizing goodness and perfection.

To make things even more confusing, Plato postulates an opposing force to nous. He labels this force ananke, which is generally translated as Necessity. It is the ‘other principle’

involved in the creation of the cosmos, in addition to nous (Plato 1937, 159-60). Ananke can be interpreted as the confinement wherein the demiurge operates. It restricts him through limitations that he himself did not create. This shows that the demiurge is not almighty like the Judeo-Christian God. Nous is a creative force, but apparently some things exist by necessity, outside the reach of nous and the demiurge. Some things just ‘are what they are’, so to say. However, nous has the primacy over ananke. Plato explains:

For the generation of this universe was a mixed result of the combination of Necessity and Reason. Reason overruled Necessity by persuading her to guide the greatest part of the things that become towards what is best; in that way and on that principle this universe was fashioned in the beginning by the victory of reasonable persuasion over Necessity. (Plato 1937, 160).

Again, going into the intricate opposition between nous and ananke greatly exceeds the aims of this thesis. What matters to my project is that at the heart of Plato’s cosmology, and his biology consequently, ‘lies the dichotomy between mind [nous] and necessity [ananke]’ (Cosans 1995, 582). Within this opposition, nous has more control over ananke than the other way around.

Everything in the cosmos exists for one of two reasons: either as part of an immaterial plan (nous), or because of material necessity (ananke). I want to propose to equate ananke with empirically observable nature. Therefore, I would characterize Plato’s dichotomy as an opposition between divinity and nature. More specifically, between divine purpose and natural necessity. Support for this interpretation can be found in the passages where Plato reacts to his predecessors, be it indirectly. As we know, Presocratic philosophers built their cosmologies around the natural elements (fire, water, earth and air) (e.g. Curd 2016, Kenny

2007). These cosmologies are refuted by Plato. He does so by reminding us that the natural elements exist by necessity. Notwithstanding that these natural factors can play an important role in any cosmology, they cannot be the formative force that drives the unfolding of the universe. They are mere accessory causes or ‘contributory-causes’ to nous (Cosans 1995, 583). Natural elements are secondary causes, but not the primary cause through which we can explain the existence of the cosmos. This primary cause can only be found in the purposive workings of the demiurge. As we have seen, the demiurge is in some sense limited by nature. However, the demiurge has primacy over nature, and therefore ananke is subdued by nous. This is the core of Plato’s critique on the Presocratics: if one wants to explain the cosmos, referring to secondary, natural causes does not provide an exhaustive account of it. The cosmos can only be sufficiently explained by referring to the immaterial, purposive causes that shape and order natural causes. Therefore, according to Plato, the most substantial deficit of Presocratic cosmologies is that they lack teleology.

Now let us turn to the subject at hand: teleology in biology. We have seen that Plato’s cosmology boils down to a contrast between purposive design and natural necessity; some things exist because of the demiurge, while other things exist because of natural necessity, outside of the demiurge’s reach. However, Plato reminds us that the demiurge has supreme authority, and that natural phenomena cannot be explained without referring to nous. To put it in more familiar terms, some things can be said to have purpose, while others cannot: they are mere mechanisms. However, mechanisms are contributory causes to purpose and therefore we cannot explain them exhaustively without employing teleology. This is the most substantial deficit in Presocratic cosmologies, as they refer to mechanisms as the ultimate explanation of the cosmos. But in doing so, they dismiss entirely the role that is played by an intentional designer. In short, the problem with Presocratic explanations is that they lack teleology. Plato makes explicit this contrast between mechanism and teleology when he turns his attention from cosmology to biology. More specifically, he turns to the workings of the human eye. Quite extensively, he describes the mechanisms that constitute and underlie vision, which he compares to mirror optics, emphasizing the purely mechanical nature of vision (Plato 1937, 154). However, this account of how light passes through the eyeball and how the consequent images are formed is not an exhaustive explanation of the

phenomenon. For the mechanistic explanation tells us ‘how’ we see, but not ‘why’. The primary cause of vision cannot be found in a mechanical account of the biological phenomenon. According to Plato, the true explanation of vision is not to be found in the underlying mechanism, but rather in the purpose it is consciously designed to serve (Plato 1937, 157). The purpose being, in the case of the eye, to aid us in our comprehension of the workings of nous. This is a line of reasoning that we are familiar with, as I have considered

the difference between ‘why’- and ‘how’-questions to some extent in the introductory chapter. It is exactly what demarcates biology from other sciences and renders it unique. For we can explain, like Plato, in detail how an organ, a biological process or an organism as a whole functions. But this explanation leaves unanswered the question why an organ, process, or organism does what it does. I deem this question an essential one. And it is a question that can only be answered, I hold, by employing a teleological explanation. The problem with Plato however is that nous and demiurge are immaterial entities that exist outside of (and in opposition to) nature. Therefore, Plato answers the ‘why’-question in the context of biological phenomena by referring to something outside of the natural order. It follows that

purposiveness in nature is imposed on it from outside. Teleology in biology is external within Plato’s framework. I reckon that this notion of teleology is quite problematic to the modern reader. Given modern scientific standards, I take it to be objectionable to make reference to entities outside of the empirically observable, physical order. Even more so when these entities are admittedly immaterial and intentional. Plato’s distinction between ‘how’ and ‘why’ is similar to Mayr’s, and his pinpointing of the paradoxical status of teleology in biology two thousand years before Mayr is remarkable. Unfortunately, his answer to the ‘why’-question is unacceptable. What is even more unfortunate in the light of our project is that Plato’s

influential reading of teleology in biology gave it a bitter aftertaste. For one of the biggest obstacles for teleology to be embraced by the scientific and philosophical community is its common association with an intentional designer and immaterial forces (e.g. Allen 2009, Mix 2016) I would argue that the seeds for this connotation were planted by Plato. But we must not give up too soon. There are other great thinkers who have shined their light upon the problem of teleology in biology. One of them is Plato’s Stagirian successor.

2.3 Internal teleology: Aristotle

Plato viewed as the main problem in Presocratic cosmologies the lack of a notion of purpose in nature. Instead, these cosmologies focussed on the mechanistic aspects of nature,

without paying attention to divine purpose. Presocratic cosmologies can be characterized as mechanistic, while Plato’s cosmology is teleological. Logically, the same can be said of his views on biology. In this chapter I will present Aristotle’s biology in a likewise manner. Like Plato’s, Aristotle’s thought can be said to have grown up in permanent opposition to the dogmatic mechanism of Presocratic thinkers (especially that of Democritus) (Driesch 1908, 144). I will present Aristotle as developing an account of teleology in biology which functions as an answer to the lack of purpose in Presocratic thought, much like I did with his

Although both Plato and Aristotle both oppose the mechanistic cosmologies of their

predecessors, their approach differs greatly. Aristotle does not refer to immaterial forces or intentional designers whatsoever. There is no external immaterial force that constitutes purposiveness in nature through the workings of an intentional designer within Aristotle’s framework.Therefore, Aristotle’s teleology is not be found in some design imposed from outside. It is internal rather than external. Purpose is not found in what is imposed from without, but comes from features within the organism. Aristotle refers to these internal features as potentiality. The textbook example on this point is that of an acorn. We all know that, given the right conditions, an acorn grows into an oak. In Aristotle’s view, the potential to become an oak lies in the acorn. He identifies an internal drive within the acorn to fulfill its potential. The acorn aims at fulfilling its purpose, so to say, which is growing into a mature oak tree. Purposes in living things can be found inside them, on basis of what they can conceivably become. Logically, they exist within the natural order, as Aristotle states: ‘the nature of a thing is its end. For what each thing is when fully developed, we call its nature, whether we are speaking of a man, a horse, or a family.’ (Pol. i.1 1252b33‐4). Therefore, Aristotle’s teleology is local and specific rather than Plato’s, which is global and universal (Zeyl & Sattler 2017). Additionally, in the passages on biology, Aristotle seems to point towards reproduction as the purpose of every organism. “For any living thing that has reached its normal development … the most natural act is the production of another like itself (...) That is the goal towards which all things strive, that for the sake of which they do whatsoever their nature renders possible. (DA ii.4 415a27‐b1)". W hen it com es to teleolog y in biology, Aristotle thus seems to show some ambiguity as to what the concrete purpose of organisms is. However, the question whether an acorn has as its purpose growing into a mature oak or generate a new one, does not really matter here. The point is that teleology in biology can be found in an innate drive that resides within every living thing. Every organism has its own specific goal, which is not dependent on a divine plan but rather on its specific internal properties. I reckon that this line of thinking about goals in living nature is a lot more agreeable to the modern reader than was Plato’s external teleology. Especially Aristotle’s comment on locating the purpose of organisms in their reproductive capabilities is intuitively legitimate given our current scientific knowledge of genes and evolution. For example, a popular line of thinking in modern biology is that if there is purpose to organisms at all, it is to be found in their being mere ‘survival machines’; passive vehicles which exist only for the purpose of transmitting their genes to the next generation (Dawkins 1976).

In the aforementioned quote from De Anima, Aristotle refers to teleological explanations as ‘that for the sake of which’. A more well-known term for these explanations is final causes

(Cameron 2010, Mix 2016)2. The final cause is only one of the possible four explanations we can provide for a given phenomenon, for:

Of a thing we can ask four different questions: “What is it?”, “Out of what is it made?” “By what agent?”, “What for”?. The four kinds of answers that can be elicited to these questions are his [Aristotle’s] four causes - formal, material, efficient, and final. Only the third type of answer (“efficient”) is causal in the modern scientific sense. (Ayala 2012, 48).

What is important to our project is that within his framework of the four causes, Aristotle identifies the dichotomy between ‘why’ and ‘how’, between teleology and mechanism, just like Plato. For formal, efficient and material causes explain how a phenomenon came to be, while final causes explain why. Again, with Plato, Aristotle contends that no natural

phenomenon can be explained exhaustively without employing both mechanical and teleological (final) explanations. He departs from his predecessor by locating the answer to the biological ‘why’-question within the natural order. For as we have seen, natural purposes can be found within organisms themselves. However, when one picks up an acorn and breaks it to inspect the inside, there will be no oak to be found. The mature oak cannot be localized within the acorn on an empirical acorn. It is only that we know (in hindsight so to say) that the acorn grows into an oak eventually. The potentiality that Aristotle presupposes in every organism therefore seems to be some sort of non-empirical vital force. It is

empirically untestable, while Aristotle maintains that it is part of living nature. Within every living being, there is some sort of unexplainable, invisible force, which drives them towards their goal3. Aristotle got rid of the purposive designer that Plato proposed. However, by positioning purpose within the natural order (namely, within the organism), new problems seem to arise. Therefore, it might be a good strategy to look for a concept of purpose that is neither external to nature in Plato’s sense, nor internal to nature in Aristotle’s sense.

2.4 Heuristic teleology: Kant

I have presented Plato and Aristotle’s reflections on teleology in biology as a reaction to mechanistic explanations of their predecessors. For Plato and Aristotle, mechanistic

2

Again, as with Mayr, it is important to note that we are talking about an explanatory factor rather than a mechanical cause in the sense of something that makes something else happen.

3

Note that it has often been remarked that Aristotle’s final cause should be taken as a “principle of intelligibility” (Ayala 2012, 48) and that “Aristotle’s telos has nothing to do with purpose” (Mayr 1992, 121). In other words, Aristotle’s interpretation of teleology is often taken as heuristic (instead of internal, as I have propounded), which is a concept that I introduce with Kant. However, for example Ruse (2017) conceives of Aristotle’s teleology as internal rather than heuristic, which I believe is more plausible. He states: “because he [Aristotle] did not think teleology merely heuristic, (...) he was rather pushed toward our second option, that teleology is in some sense a function of, necessitated by, a kind of underlying end-directed force or forces (...). The main point is that when we talk of purpose [in Aristotle], we are in some sense talking of something objective rather than just subjective.” (Ruse 2017, 3:28). With Ruse, I take Aristotle to presuppose an objective, unempirical vital force in nature in his concept of teleology, rather than a subjective mode of intelligibility, which I reckon is a much more modern way of thinking that originated with Kant and not with Aristotle.

explanations alone cannot be exhaustive when it comes to biological phenomena, especially organisms. Teleological explanations must be employed in order to provide a

comprehensive explanation in biology. However, we have also seen that this stance is quite problematic for the modern reader. In this section, we will examine Immanuel Kant’s view on teleology in biology, which is, as we will conclude, more in line with modern scientific

thought. For we have presupposed that modern science mainly relies on mechanistic explanations. Kant appreciates this, and in contrast to Plato and Aristotle he attempts to reconcile mechanism with teleology, instead of the other way around. Plato and Aristotle saw mechanistic explanations as flawed and argued for teleological explanations as the most appropriate model for biology. Kant on the other hand strongly believes that mechanism is the most suitable mode of explanation in science, while teleology cannot be exhaustive. In this section we will examine his interpretation of teleology in biology. Finally I will conclude that Kant’s approach to teleology in biology is most fruitful in our project of ‘rehabilitating’ teleological language in biology.

As I have mentioned, for Kant, mechanical explanations are the ideal in biology as well as all the other sciences. They can provide us with objective knowledge of nature, in the sense that for example Newtonian physics does. In contrast, teleological explanations cannot provide us with objective knowledge of nature (Van den Berg 2014, 262). This does not mean however that Kant closes the door entirely for teleology. There is a role left for

teleology to be played, as mechanistic explanations do not suffice for organisms. This point can be clarified by delving a little deeper into the definition of mechanism. Kant’s conception of mechanism is commonly interpreted as a relation between a whole and the properties of its parts (Van den Berg 2014, 55). In this interpretation, mechanistic explanation consists in explaining the whole (of a given phenomenon) in terms of its parts and the causal relations between them. The parts posses these properties on their own in the sense that they are independent of the other parts (ibid.). Kant elucidates his view when he mentions a watch, which is the prime metaphor for talking about mechanisms since Descartes: “In a watch one part is the instrument for the motion of another, but one wheel is not the efficient cause for the production of the other: one part is certainly present for the sake of the other but not because of it” (Kant 2000, 24).

Mechanical explanations thus employ part-whole schemes where the parts are causally connected to each other but not mutually dependent. They influence each other, but they do not depend on each other for their existence. As I have mentioned, according to Kant this mode of explanation is the ideal for science, biology included. However, Kant identifies a problem in biological explanations. For this type of mechanical explanation does not apply to

organisms. Take for example the eye. With Plato we have seen that, for the most, the workings of the eye can be explained in mechanistic terms. We have also seen that this mode of explanation does not explain the eye completely. For the eye is useful to the organism as a whole, and in this sense brings about survival and reproduction. So the function or purpose of the eye is left untouched by a mechanistic explanation of the eye. Both Aristotle and Kant saw this. Kant emphasizes something rather peculiar however. In bringing about survival and reproduction for the organism, the eye (indirectly) produces another eye (Ruse 2017, 3:45)! Therefore, it seems to be the case that parts of biological wholes are not only causally dependent upon each other, but produce each other as well. He is therefore forced to conclude that organisms, and everything contained in them, are

somehow both cause and effect of themselves (Steigerwald 2010, 292, Van de Vijver et al. 2003, 105). Luckily, Kant provides us with an example to clarify this rather confusing statement, by pointing out three distinct features of a tree (Kant 2000). Firstly, he mentions reproduction. Within a specific genus, trees beget new trees through reproduction. From the perspective of the genus, trees are thus both the causes of themselves, but also effect: the genus produces and reproduces itself perpetually. Secondly, he draws our attention to individual trees. Their growth is a special kind of generation, because the matter out of which the tree is built up contains qualitative features that nature external to it does not. To clarify this point anachronistically, we can refer to organic chemistry: every component of a tree is largely built up out of carbon compounds, but the quality of ‘being a tree’ is not intrinsically contained in these compounds. Kant explains this point the other way around: it is

impossible for us to recreate wood by using the building blocks that it is made out of. It follows that trees are made up out of their own products, so that their effects are their causes at the same time. Finally, he zooms in to the parts that constitute a tree. Every root, twig and leaf is a product of the tree while it sustains it simultaneously. Additionally, every part is reciprocally dependent upon the system as a whole, while it maintains some form of autonomy as well (for example, Kant reminds us of the practice of grafting, where parts of plants are removed and grow into new plants).

We can conclude that mere mechanistic explanations do not suffice for organisms. For organisms cannot be explained merely as wholes that are constituted by their parts. Parts of the organism exist for the sake of each other and for the sake of the whole, but additionally exist because of each other as well. The parts of an organism are reciprocally dependent of each other for their existence. A watch can be explained by referring to the wheels and clogs it is made up out of, but the wheels and clogs do not exist because of each other (they are made by an external watchmaker, and do not reproduce themselves). An organism however is an entirely different story. It recreates itself and its parts create other parts. This is why

mechanistic explanations are not satisfactory in biology. For mechanistic explanations (at least in Kant) explain wholes by identifying the independent properties of the underlying parts (Van den Berg 2014, 55). However, when the parts have properties that are dependent of other parts and of the whole, which is the case with organisms, mechanical explanations cannot suffice, as they leave certain features (namely that the parts are interconnected through dependency) unexplained. Therefore, we need a different kind of explanation in order to fully understand organisms. We need an explanation that can account for the reciprocal, self-organizing nature of living things. We need teleological explanations, which account for these features by employing goals or goal-directedness as an explanatory factor. Kant identifies organisms as natural purposes (Steigerwald 2010, 294, Van den Berg 2014, 91). We can only understand the unity of their reciprocally related parts by appreciating the fact that organisms are purposeful. Therefore, we can only understand biology by employing teleology.

Ruse (2017) proposes that Kant’s interpretation of teleology can be viewed as heuristic. Purpose in nature is not imposed on it by an intentional designer (Plato), nor is it part of the natural order (Aristotle). Teleology is merely a figure of speech, a mode of understanding and talking about the peculiar features of organisms which fall outside of mechanical explanations. I take this view on teleology to be most in line with our modern intuitions regarding decent scientific practice. For purposes in nature have an unscientific status, as Kant rightly concluded. At least organisms do not have intentional purposes like human beings do (or at least not all organisms). Nor is there an external immaterial designer who infuses the natural order. A purposive force which is part of every organism is equally

unlikely. However, we cannot talk about organisms without employing teleology. Kant seems to circumvent this paradox by downplaying or degrading biology as a science (Ruse 2017, 3:49). This is not an option for me. I believe that biology is a highly important and unique science and is respectable as such. In order to ‘protect’ biology from losing its standing, we will use Kant’s heuristic interpretation of teleology, stripped off its contempt for biology so to say. We will employ his approach to teleology without following his conclusion that biology is a second class science or anything along those lines. For a commitment to teleological language does not entail an ontological commitment to goals and purposes in nature: we need these concepts in order to talk sensibly about organisms and demarcate biological phenomena from non-living ones. This does not mean however that we believe that intelligent designers and immaterial forces are part of the natural order. It is our aim to provide a conceptual framework for biological language. The unraveling of the fabric of empirical reality is a task that should be left to the scientist. Therefore, with Kant, we take a definitive turn from the ontological level of enquiry to an epistemological one. Where Plato

and Aristotle were concerned with proving the existence of purposes in nature, with Kant we will interpret purposes in nature as a mode of understanding from now on. Teleology is not a mode of existence, but a mode of intelligibility. Purpose is not external to nature in the form of a divine plan, nor is it internal as a residing force of potentiality. Purpose is merely inside of us, as an apparatus within our inquisitive minds. It functions as yet another tool for understanding the (living) world around us. However, in contrast to Kant, I would contend that biological discourse might be different from ‘regular’ causal-mechanistic science, but not inferior.

Let us summarize what we have learned about Kant’s view on biological explanations: organisms can be explained in mechanistic terms but only to a certain degree. For they differ from ‘regular’ mechanisms in that the parts are not only causally connected to each other but mutually dependent as well. Parts of an organism constitute each other and reproduce themselves. In this sense, they are both cause and effect of themselves. Another way of describing this is that organisms are self-organized, where ‘regular’ mechanisms show organization but not self-organization. It seems that at least in the case of organisms, mechanistic explanations alone do not suffice to provide an exhaustive explanation. For we have seen that the structure of organisms, the reciprocal dependency of their parts and their self-organizing and self-creating features cannot be explained by mechanism alone. We need to conceive of organisms as goal-directed natural purposes in order to capture their full complexity. Therefore, Kant is forced to allow for teleological explanations in biology,

whether he likes it or not. For mechanical explanations of living nature do not suffice, and therefore the employment of a teleological method is justified (Van de Berg 2014, 100). Let us end this chapter with a few concluding remarks. We have established that Kant’s ideal scientific explanatory model is mechanistic. Therefore, teleological explanations have an unscientific status within his framework. While the nature of organisms demands teleological explanations, Kant does not seem to be positive towards assigning scientific status to these claims. Additionally, Kant points towards the connection between teleology and normativity. For “to construe a natural object as purposive is to make the normative claim (...) that there is some way the object ought to be” (Van den Berg 2014, 104) (my italics). For example, when we explain the eye teleologically, in the sense of ‘the eye serves to see’, we make an essentially normative claim. For within this teleological explanation, “we judge that it [the eye] ought to be suitable for seeing” (Van den Berg 2014, 104) (author’s italics). At least since Hume, it is commonly accepted in philosophy that nature cannot tell us how things ought to be, only what things are. Normativity may form yet another obstacle for legitimizing teleology in biology. We will assess this problem to a greater extent in chapter four. For now, it suffices to note that Kant’s emphasis on the normative character of teleological

explanations once again shows his heuristic or subjective notion of teleology: for norms are not to be found in nature but come only from our minds (at least this is what Hume showed). The teleology (and consequent normativity) we employ to explain organisms is therefore not intrinsic to them (the objects of explanation), but a product of our minds (the subject that does the explaining) (Van den Berg 2014, 104). Lastly, Kant identifies yet another problem with teleological explanations. In addition to the fact that they do not sit well with his ideal model of mechanistic science and that they convey normative judgments, Kant thought that organisms as natural purposes display reciprocal causality. For as we have seen, they parts of an organism (and the organism in it totality) are both cause and effects of themselves. Within this reciprocal causality, “A causes a future effect B, while conversely that future effect B causes A” (Van den Berg 2014, 102). The future effect B causing A is what is commonly referred to as backward causation. For in mechanistic explanatory models, causes precede their effects in time. However, as Kant rightly notes, in teleological explanations future effects (in the form of goals) seem to precede past causes. This is a reversal of what is commonly identified as proper causation and is quite problematic. One could argue that backward causation (more specifically, reciprocal causation, which goes ‘both ways’) is what vexed Kant about organisms. For we have seen that the reciprocal structure of the causal relationships between the parts of an organism is what leads Kant to conclude that organisms should be thought of as natural purposes. Reciprocal causation is what deems organisms unfit for mechanical explanations. We will use this as a starting point for the next chapter. With Kant’s heuristic approach, we may have laid the foundations for a scientifically acceptable interpretation of teleology (namely, one that does not refer to designers or vital forces). However, while the historical problems may seem to have been solved, a more structural problem now seems to arise. For if teleology makes reference to the future, it might not ever be able to achieve scientific status. This would force us to conclude, just like Kant, that biology is some sort of second class science. Fortunately, five years after Kant’s death, a man was born who would change biology forever and, as we will see, provide a tool to possibly reconcile teleology with science. I am talking about Charles Darwin, and the tool that has been put to use by philosophers to further analyze teleology in biology is evolution through natural selection.

3. Evolution & the Etiological Approach

3.1 General introduction

In the previous chapters, I have attempted to show that teleology is bound up with biological discourse on a fundamental level. However, teleological explanations prove problematic when it comes to their scientific standing. Therefore, it is important that we interpret teleology in a specific way. Teleological explanations should not be taken as referring to intentional designers nor should goals and purposes be taken as part of the natural order. Teleological explanations in biology are acceptable if and only if we take them as a mere mode of

understanding, as a figure of speech so to say. Teleology is a metaphor, but a metaphor that biology cannot do without: this is the heuristic interpretation that Ruse ascribes to Kant (Ruse 2017, 6:23). From the outset, a science that is highly dependent upon metaphor does not seem legitimate. Kant arrived at this exact conclusion. Moreover, teleology seems to employ backward causality, which does not fit with the mechanistic ideal of science. However, modern biology is a well-established and respectable science. In fact, out of the ten most cited academic papers of all time, seven are on (methods in) biology. This provides testimony to the vast amount of researchers that are working in this hugely diversified scientific field. Therefore, we cannot and we should not belittle biology and its methods. On the contrary, I would argue, we should cherish the groundbreaking discoveries and insights that the life sciences have provided us with. In order to do so, we need to find an acceptable conceptual ground to justify teleological modes of discourse in biology. The biggest obstacle to justifying teleology in this chapter will be accounting for the problem of backward

causality.

We will do so in this chapter by examining the work of two prominent practicing biologists and influential philosophers of science: Ernst Mayr and Francisco Ayala. Both authors

identify the problem of teleology just like Kant did. In solving it, these authors find solutions in a biological concept that was not available to Kant: the theory of evolution. We will see that evolutionary biology can provide a scientific grounding for teleological language. Before we move on however, we will need a basic understanding of evolutionary theory. Therefore, in the first part of this chapter, I will present a very preliminary summary of classic Neo-Darwinism.

3.2 Evolution

Since Darwin, and especially after the formulation of what has been named ‘the modern synthesis’ during the first period of the 20th century, biological discourse is primarily concerned with and shaped by evolutionary theory (Okasha 2009, 7, Ayala 2012, 35). If

biology is to be viewed as a unified science at all, Darwinism is the central unifying concept, comparable to Newtonianism in physics (Okasha 2009, 2). Not only has it left its mark on the life sciences, evolutionary theory branched out to various other scientific fields as well; economics, psychology, computation and anthropology are among them. Understanding evolution is therefore essential to understanding and analyzing biology.

Evolution is a complex process which consists of a vast set of interrelated systems, but the general framework of it can be discussed in a straightforward manner. Darwin was, like Plato, Aristotle and Kant, fascinated by the intricate design of the living world as he

encountered it. His theory of evolution would successfully attempt to explain “biology’s two great explananda” namely adaptedness and diversity (Walsh 2006, 774). Adaptedness is the empirical given that every living organism one encounters, seems to be exceptionally well-equipped to cope with its direct biological environment. Think of the giraffe’s extraordinary long neck, which provides opportunity to reach nutritious leaves in the treetops of acacias. Another testimony to nature’s adaptive capabilities can be found in various species of deep-sea fish. The black deep-seadevil for example owes its rather bizarre body composition to its ability to withstand extreme pressure in the depths below. Prime examples can be found in the group of microbes commonly referred to as ‘extremophiles’: these micro-organisms have adapted to the most extreme conditions recorded on Earth and have been discovered in extremely arid, saline or acidic environments, withstanding temperatures ranging from -15° to 120°C. The second object of evolutionary explanation is diversity. This is the empirical given that different organisms seem to be arranged into distinct kinds or types. In other words, the fact that we can sensibly state that one organism differs from another one, makes that they are intuitively categorized. Birds intuitively form a demarcated species, as do reptiles and rodents. Even small children can identify these naturalized categories, as every organism that is identified as having ‘two legs, wings, feathers and a beak’ can be

recognized as belonging to the categories of ‘birds’. Obviously, there are many examples of organisms that are not so easily categorized, like the platypus. However, this is a mere testimony to the immense diversity in living nature, of which the thorough systematization was first endeavored by Linnaeus. This undertaking might not ever see its completion, as numerous new species are still discovered every year. In summary: ‘evolution provides a scientific explanation for why there are so many different kinds of organisms on Earth and gives an account of their similarities and differences’ (Ayala 2012, 35). Nothing in modern biology makes sense without it.

Darwin was not the first to notice the adaptive and diverse nature of the living world. However, the novelty lay in the explanation Darwin proposed for these phenomena:

evolution through natural selection. With Darwin, adaptation and diversity in nature could now be explained without referring to immaterial life forces or an intelligent designer. One could argue that Darwin made biology into a fully fledged modern scientific practice. In a nutshell, his most important discovery was that through the reproduction of organisms, certain traits are transferred onto future generations. The transferring of traits, which leads to variety in populations over time, is what we call evolution. Essentially there are two major forces that drive evolution: mutation and natural selection. Mutation is the basic source of random hereditary variability, while natural selection is the main formative force of organized hereditary material (Ayala 1970, 20). Evolution is a process that can be described as

“opportunistic”, be it stripped of any anthropomorphic connotation the word entails (Simpson 1949, 160). It is opportunistic in the sense that variations, both in hereditary material and in the environment, provide natural selection with opportunities of perpetuating the most favorable outcome (Beatty 2008, 9). The interactive combination of random variation and formative enhancement, set against an ever-changing environment, comprises the essential structure of evolution. In a nutshell, these two processes cause evolution (the environment is a hugely important third factor, although it is independent of the evolutionary process as such). It is random mutation in genetic material that causes differences between individual generations and natural selection selects for these changes, effectively determining the direction of evolution.

Let us turn to an example in the form of a familiar organism: the giraffe. Imagine a

hypothetical moment in biological history where a herbivore with a ‘regular’ sized neck (the okapi, a close relative of the giraffe, actually has a ‘regular’ sized neck and a similar animal would be suited to picture in one’s head) finds itself in a barren tundra. Environmental factors cause sources of nourishment to deteriorate.At one point, the leaves on top of acacia trees become the only viable source of food for this particular animal in this specific environment. This is the moment where the two-fold system of evolution kicks in. Suppose that this ‘regular neck size’ mammal has offspring. As mutations cause intergenerational differences in organisms, one of the descendants is, by chance, born with a slightly longer neck than its siblings. This provides our ‘slightly longer neck’ individual with a slight advantage over its relatives in reaching its main source of nutrition. The individual has a slightly higher chance of surviving and, consequently, has a slightly higher chance of living long enough to find a mate and produce its own offspring. It is therefore more fit to reproduce. Accordingly, the mutation causes a relative increase in reproductive fitness. The offspring produced by our ‘slightly longer neck’ contains, in turn, a new individual with an even longer neck. This ‘even longer neck’ individual has higher chances of surviving and procreating, and so forth. This feedback loop, initialized by a random mutation and perpetuated by reproductive fitness, is

what we call natural selection4. Note that natural selection as a concept inherently contains a number of presuppositions. The most important of those are that entities in a population are able to reproduce, and that in doing so they transfer heritable traits which vary. (e.g.

Dawkins 1976, Lewontin 1970). It is commonly agreed upon that non-living entities in the ‘primordial soup’ (the term which is often deployed for the state of planet Earth before there was life) would have somehow needed to develop these features in order for evolution through natural selection (and thus life) to come into being. These three factors are not exhaustive however, and others can be formulated. So much for our ‘Darwinism 101’. We are now ready to delve further into the work of post-Kantian authors, who are heavily embedded in evolutionary theory.

3.3 Francisco J. Ayala

As I have mentioned before, biology can be said to have become a modern science only after Darwin. One supporter of this proposition can be found in Spanish-American evolutionary biologist and philosopher Francisco J. Ayala, who writes regarding Darwin:

One of his greatest accomplishments was to bring the teleological aspects of nature into the realm of science. He substituted a scientific teleology for a theological one. The teleology of nature could now be explained at least in principle, as the result of natural laws manifested in natural processes, without recourse to an external Creator or to spiritual or non material forces. At that point biology came into maturity as a science. (Ayala 1970, 2).

Ayala’s analysis of teleology in biology relies heavily on etiology. This means that Ayala tries to define teleology as backward-looking. The main point of his approach is that the historical coming-to-be of a biological phenomenon is what defines it as teleological. In this section, we will delve into Ayala’s take on the problem of teleology in biology, and see whether his interpretation leads the way to the solution that Kant could not find. From the outset, as I have stressed throughout this thesis, Ayala notes that teleological explanations sit

uncomfortably with modern readers and that ‘the concept of teleology is in general disrepute in modern science’ (Ayala 1970, 8). In the previous chapter, we have identified two historical reasons for this disrepute: the connection of teleology with a notion of an intelligent designer (which we traced back to Plato) and a notion of immaterial life forces (which we traced back to Aristotle). Ayala however identifies a third notion which brought teleology into discredit: backward causation. For he believes that the main reason for its negative reputation is ‘that the notion of teleology is equated with the belief that future events are active agents in their

4

It goes without saying that this picture of evolutionary theory is greatly oversimplified, as natural selection takes place on a vast time scale, across many consecutive generations. The same goes for alterations in the environment. Also, there are numerous other sources of intergenerational variability, sexual recombination being one of the most prominent.

own realization (Ayala 1970,8)’5. Historically, at least since Hume, causation is taken as causes preceding their effects in time. We have seen that Aristotle identified this form of causation as “efficient”, in the sense of that what brings something about. The efficient cause is the agent that both causally and temporally precedes its effect. Ayala recognized

Aristotle’s efficient cause as the only one out of four that is causal in the modern scientific sense (Ayala 2012, 48). Any other temporal sequence of cause and effect would be neigh unimaginable. Therefore, the notion of backward causation is seen, at least historically, as involving ‘either a contradiction in terms or a conceptual impossibility’ (Faye 2018). Others have referred to this as ‘the argument from nonactuality’ (Walsh 2008, 5). For teleological explanations employ goals, which can by definition be identified only when they are attained. However, the biological phenomenon that is explained teleologically, precedes its goal temporally. For example, we can sensibly state that the premature hand of a human foetus is made for grasping. We can do so only, so to say, in hindsight, as this goal has not yet been attained. Moreover, if by some unfortunate event the foetus in question would not have been born at all, we will still explain the function of the hand in this manner. In this case however, the goal has not been and will not ever be attained. The goal that is employed by the teleological explanation is unactualized. In other words: ‘means precede their ends, but in teleology, ends explain their means’ (Walsh 2008, 5) (author’s italics). Teleological

explanation by definition “makes reference to the future” (Ruse 2016, 101). Where we have identified historical arguments against teleology (with Plato and Aristotle), Ayala seems to view this systematic or structural argument as the main reason for general aversion against teleological language in biology. And this is why, for him, an etiological approach to teleology is the way to go.

The solution he provides can thus be presented as an attempt to circumvent or overcome the problem of backward causation. His interpretation of teleology starts with the observation that “teleology arose most probably as a result of man’s reflection on the circumstances connected with his own voluntary actions” (Ayala 1970, 8). This reminds the reader of the example I put forward regarding me writing this thesis: we can analogously extend our understanding of human goal-directed behaviour to other species. We can take this analogy to extremes, e.g. plants or bacteria, as long as we do not presuppose intention as we do with human behaviour. Now, in an attempt to counter both the problem of backward causality as well as that of intentionality, Ayala proposes the following definition of teleology: teleological explanations require a feature to contribute to the phenomenon in such a way that this contribution is the explanatory reason for the existence of the feature in the first place (Ayala

5

1970, 8, Ayala 2004, 249). He stresses that the contribution “must be the reason why the feature or behavior exists at all” (Ayala 2012, 39-40). Taken at face value, this definition does not seem very informative or clear. I found illuminating the remarks on this point made by Karen Neander (1991). Similar to Ayala, Neander defines the etiological approach as claiming “that what counts as a function [purpose, goal] of a trait is determined by that trait’s history” (Neander 1991, 459). In other words, we can say that an organism’s trait has a purpose or a goal when it is historically inclined to achieve this goal. She describes the etiological approach as explicitly referring to future effects of a given trait, by implicitly

referring to the evolutionary process that happened in the past as an explanation for that trait (Neander 1991, 463). She agrees with Ayala that backward causation is an obstacle to overcome in order to provide teleology with a scientific status: “we do not understand

teleological explanations correctly, (...) unless we understand that they are not only explicitly ‘forward-looking’ but also implicitly ‘backward-looking’, and it is only in virtue of this implicit looking back to prior causes that teleological explanations are explanatory” (ibid.). Let us return to the example of the unborn foetus’ hand. I have introduced the problem of backward causation by showing how the purpose of the hand may not be fulfilled at all if the baby was not to be born. The etiological approach counters this by stating that it is not the future goal as such which is constitutive for the structure of teleological explanation. The future goal of the hand is determined by the historic process of evolution. The purpose of the hand, which is grasping, has proved beneficial to our species during the process of natural selection. Therefore, grasping is the purpose of the hand, regardless of whether this purpose is actualized. The outcomes of past contributions to reproductive fitness determine the future ends of traits and processes within the etiological approach to teleology.

Returning to Ayala, I would argue that he adds an epistemic condition to Neander’s

definition. Not only is the future goal-oriented trait determined by the past selective process, it must be the explanatory reason for it as well. This provides testimony to the heuristic viewpoint that is entertained by Ayala (and myself) when it comes to teleological explanations. Teleology is not only a convenient figure of speech, some (if not most) biological phenomena cannot be explained without it. For example, we can explain the shape of a mountain as a result of geological processes, namely tectonics or volcanic forces combined with erosion. In a similar manner, we can explain the shape of a heart as a result of the physiology of the body, namely the cardiovascular system which consists in the distribution of blood, providing the entire body with nutrients and oxygen. It is the purpose of the heart that provides it with its reason for existence in the first place. While the mountain is shaped by geology just like the heart is shaped by physiology, it makes no sense to state that the mountain maintains this shape for a certain end, like for example skiing (Ayala 2012,

39). In contrast, we can reasonably say that the heart received and maintains its shape in order to fulfill its purpose: pumping blood around. It was designed to do so through the process of natural selection and its future purpose is defined by it. Let us return to our giraffe-example to clarify this further. We have already painted a picture in our minds of the giraffe’s (hypothetical) early ancestor: an okapi-like creature. It is safe to say that there was nothing in this early ancestor’s physical constitution that would have led to its descendants having a neck that measured over two meters in length. As we have seen, it is a long chain of stochastic events, consisting of mutations and environmental factors, guided by natural selection, that separates the okapi-like ancestor from the modern-day giraffe. However, Ayala stresses that this feature came about precisely because of the purpose of reaching acacia leaves. The explanatory reason for the longevity of the giraffe’s neck is the end it serves (Ayala 2012, 43). As Neander notes, the causality within the etiological approach to teleology remains to have a forward-looking character, as the evolution from the okapi-like ancestor to the giraffe is explained by reference to the future goal of reaching leaves. However, the explanatory power comes from a backward-looking structure in the sense of regular causality. For the process of evolution can be traced back into history, and therefore provides an explanation where ‘regular’ (efficient) causation is employed. It is therefore scientifically acceptable. And not only is the causality of teleological explanations both forward-looking and backward-looking, Ayala adds that there is no other way to explain evolutionary features than by employing teleology. This leads him to conclude that teleological explanations in biology are not only appropriate but necessary as well (Ayala 2012, 49). Note that we now arrive at a similar conclusion as we did with Kant. For with Kant, we have already seen that the causality of teleological explanations is reciprocal (both

forward- and backward-looking). The difference here is the introduction of evolution. For Kant could not solve this peculiar character of teleological causality. With Ayala (and Neander), we have now seen that evolution can be employed to dissolve this seemingly perplexing structure of teleological causality, by scientifically grounding it in evolution. Evolution provides future goals with a history, so to say.

3.4 Ernst Mayr

Mayr also attempts to rid teleology of its backward causation. He does so by employing the notion of program. Yet, we have to briefly discuss his general framework before we can dive into this concept. He commences his defense of teleological explanations by distinguishing the different uses this word has been put to. He is quite unambiguous in his statement that “those recent books and papers [on the problems of teleology] are quite useless which still treat teleology as a unitary phenomenon” (Mayr 1992, 121). Up until now, we have used various meanings of teleology synonymous: for terms like ‘end-directed’, ‘purposive’,