Top-Down Causation (Part 1): Mechanisms and Their Levels

Tweet

This post is part of my series on the work of the philosopher Carl Craver. The series looks at the nature of neuroscientific explanation. For an index, see here.

Over the next few posts, I will be looking at the following article:

Craver and Bechtel "Top-down Causation without Top-down Causes" (2007) 22 Biology and Philosophy 547.

The article deals with the problem of top-down causation. This is the idea that the higher-level, abstract properties of a mechanism can causally affect the lower-level components. If you have ever wondered whether your thoughts cause your neurons to fire, or whether your firing neurons cause you to think, then you have wondered about this problem.

The difficulty with top-down causation is that it seems ontologically spooky. Craver and Bechtel offer a qualified defence of the concept. They suggest that what is referred to as top-down causation is a real phenomenon, but that it is best explained in terms of the constitutive relations between the levels of a mechanism.

There is a lot of technical jargon being employed here. In this first post, I will try to cut through the jungle of jargon.



1. What is a Mechanism?
The first thing we need to know is: what exactly is a mechanism? This question has been exhaustively answered elsewhere on this blog. In the interests of brevity, only three properties of mechanisms are described here:

• A mechanism is a spatially and temporally organised collection of entities and activities.
• Mechanisms are affected by and have effects on "things".
• The individual entities and activities of the mechanism do not have the same effects in isolation from the mechanism.

An example might help to flesh-out these ideas. Take an MRI scanner. It maps the spatial and temporal distribution of certain molecules in the human body. So, it is a mechanism with a particular cartographic capability.

An MRI scanner is a long tubular construction consisting of three coils. The first coil generates a magnetic field, the second corrects for inhomogeneities in the magnetic field and produces fields that vary along three dimensions, and the third sends out radiofrequency pulses and records the echoes that are returned. When put together, these coils have certain effects on the hydrogen atoms in the human body. Effects that allow us to build-up a 3-d image of the tissues in the body.

On their own, the components would not have these capabilities; together, they do. Thus, we can begin to see how talk of top-down causation is so common.

2. The Levels of a Mechanism

Of course, to fully appreciate the attraction of top-down causation we need to appreciate how there can be different "levels" in a mechanism. The human eye provides a beautiful illustration of this phenomenon.

The eye is mechanism that transduces (i.e. changes) a photonic signal into a neural or electrical signal. This is the overall change that is brought about by the mechanism. The overall mechanism is itself constituted by separate mechanisms. First, there is a mechanism for collecting the light; then a mechanism for focusing the light; then a mechanism for projecting the light onto the retina; and then a mechanism in rod and cone cells for absorbing the light and converting it into a neural signal.

There is a well-known mechanistic story to be told about what happens within the rod cells as well. The absorbed photons change the conformation of rhodopsin (a protein found in rod cells). This stimulates G-proteins, which in turn set-off a biochemical cascade leading to the hyperpolarisation of the cell. If you would like to get more detail on this story, I suggest the following video (long, but worth the effort):

The eye is thus to be thought of as hierarchical nest of mechanisms. The question to ask is: what distinguishes higher levels from lower levels? The informal way of looking at it is that higher levels are spatially and temporally extended, when compared with lower levels, and that the higher levels are aggregated from the lower levels.

We can turn this into a more formal definition:

 An item (X) is at a lower level than an item (S) if and only if X is a component in the mechanism for some activity (V) of S. And X is a component in a mechanism if and only if it is one of the entities and activities organised such that S Vs.

So, G-proteins are at a lower level than rod cells, because G-proteins are part of the mechanism that allow rod cells to send neural signals. And rod cells are in turn at a lower level to the eye because they are part of the mechanism that allows the eye to transduce photons into neural signals.


That's it for Part One. In Part Two we will consider exactly why interlevel causation might be thought problematic and sketch Craver and Bechtel's solution.

Oppy on Moral Arguments (Part 2): The Argument from Virtue

Tweet

This post is part of my series on Graham Oppy's discussion of moral arguments. For an index, see here.


The Argument Stated
The second argument that Oppy considers is the argument from virtue. This is an odd one: it argues from the alleged moral superiority of believers to the truth of their beliefs. A formal version would look like this:

• (P1) People who do not believe that there is an orthodoxly conceived monotheistic god are morally inferior to people who do believe in such a god.

• (P2) People should always choose that which is morally superior over that which is morally inferior.

• (C1) Therefore, people ought to believe that there is an orthodoxly conceived monotheistic god.

• (P3) Anything that one ought to believe is true: one cannot have an obligation to believe falsehoods.

• (C2) Therefore, there is an orthodoxly conceived monotheistic god.

This is an argument of two halves. The first half, up to C1, tries to establish that we have an obligation to believe in God. This is called a "doxastic" obligation. It is distinct from a moral obligation. The former is about beliefs; the latter is about conduct.

The second half, from C1 to C2, tries to establish the truth of our doxastic obligations.

Now, we must ask: is the argument any good?

Assessment

The problems begin with P1. It is not clear how the moral superiority of believers is established. One possibility, beloved of evangelical Christians, is that the belief is itself morally virtuous, irrespective of its effects on our conduct. If you've ever heard someone saying that the acceptance of Jesus is the ultimate moral act, you have come across this claim.

But this is a pretty lousy claim. Assuming that objective values cannot rely on God for their existence (something explored in part one), we must further assume that God is non-moral (although he may have moral properties). It is difficult to see how believing in something that is intrinsically non-moral can be morally virtuous. Only the related properties would have that power. And so at best, we should acquire belief in the properties that make God morally virtuous.

In any event, even if conduct (as opposed to belief) is the test of moral superiority, it is not at all clear that believers are morally superior to non-believers. Any fair trawl through the historical record makes it pretty clear that believers and non-believers are equally capable of morally atrocious acts. Compiling a balance sheet comparing believers and non-believers would be difficult and not very illuminating.

Even if we accept P1, there are good reasons to reject the argument as a whole. The main problem comes from it's willingness to jump from moral obligations to doxastic obligations. Again, moral obligations are about conduct, doxastic obligations are about belief. There seems to be no contradiction in stating that one could have a moral obligation to believe something that was not true.

Finally, there is no reason to assume that a doxastic obligation to believe in something would actually establish the truthfulness of the belief.

Oppy on Moral Arguments (Part 1): The Argument from Objective Values

Tweet

This post is part of my series on Graham Oppy's discussion of moral arguments for God. For an index, see here.

The first argument that Oppy looks at is the argument from objective values. It usually has something like the following form:

• (P1) The Moral Law, has a real, absolute, objective existence.

• (P2) If the Moral Law has a real, absolute and objective existence, there must be a Mind from which the Moral Law is derived.

• (C1) (Hence) There must be a Mind from which the Moral Law is derived.

Oppy takes this version of the argument from Rashdall (1907),* it is also similar to the argument offered by C.S. Lewis.

1. Terminological Prelude

Before we can even begin to assess this argument, we need to be clear about the meanings of the terms being used. First, "the Moral Law" is obviously some sort of prescriptive law that tells you what you should and should not do ("You should do X"). Another way of putting this is that the moral law gives you reasons-for-action. This makes it distinct from a descriptive law, which simply offers a formal description of the relationship between you and your actions (e.g. "You do X, because of Y").

Next, we must ask what it means to say that the Moral Law has "real existence". I take it that this simply means that the moral law exists, i.e. it is not fictional. Thus, there is something in existence that gives us reasons-for-action.

Of course, it is not enough that there exists something that gives us reasons-for-action. That something must be "absolute". This means it must hold true in all times and places: it is not relative to the needs of particular individual, society or historical moment.

Finally, it must have an objective existence. This seems to mean that it is not observer-dependent; that it is not a matter of subjective opinion; and that it somehow inheres in the fabric of reality.

2. Analysis

Now that we have some idea of what the argument means, we can assess its merits. We must question both of its premisses. We begin with P1. It is open to us to reject P1. There are two ways in which we can do this.

First, we can argue that it is impossible to satisfy all three requirements ("reality", "absoluteness" and "objectivity") at the same time. We can say that an objective and absolute law would fail to give us reasons-for-action because what motivates us is (a) subjective and (b) context-dependent. See my post on realism for more.

Second, we can provide an alternative account of our moral beliefs. In other words, we can explain why we might think there is a moral law, even though there really isn't. For example, we might feel bound to particular rules because (a) we are projecting our own emotions onto reality or (b) we are blindly following the morality of our culture.

Let's assume that we accept P1. If we do, we must ask two crucial questions about P2:

1. Is there a plausible non-theistic theory that accounts for the reality, absoluteness and objectivity of the Moral Law?
2. Does theism actually account for the reality, absoluteness and objectivity of the Moral Law?

In relation to the first question, we must ask how moral facts and properties could fit within our worldview without theism. There are plausible non-theistic theories that allow for a Moral Law. For instance, we could take a Hobbesian position and argue that objective reasons-for-action can arise from contracts for mutual advantage; or we could take a Kantian position and argue that the mechanism of practical reason automatically gives us objective (and absolute!) reasons-for-action. These are both theories that allow for moral facts to supervene on natural facts. To counter this, the theist would have to start shouting "No! no! no! Only God can account for a Moral Law!". But this would be a classic instance of begging the question.

In any event, it is not at all clear that God does a better job of accounting for a Moral Law. The problem is that theists want God to make a difference to morality. For them, it cannot be the case that the Moral Law is a necessary truth: it must be contingent upon God's existence. This forces us to imagine that there are many possible worlds in which the Moral Law could have been different. For instance, there must be a possible world in which God could have made the Holocaust morally permissible (or even obligatory). But this destroys the idea of morality: even if such a world was possible, indeed, even if we are now living in that world, we would have no reason to follow this Moral Law.

In short, making morality dependent on the contingent choice of God makes it impossible. If there are genuine moral facts, then these must be necessarily true, or true in all possible worlds.

* "The Moral Argument for the Existence of God" reprinted in Hick (ed) Classic and Contemporary Readings in the Philosophy of Religion (Englewood Cliffs, NJ: Prentice-Hall, 1964). 

Oppy on Moral Arguments (Index)

Tweet

I seem to have an obsession with arguments linking morality to religious belief. I'm not entirely sure why this is since moral arguments tend to be pretty weak. I think my reasoning is something along the following lines:

Although there are good, independent reasons for rejecting the existence of God, moral arguments have a particular practical importance. This is because moral purpose is central to the religious life: unless God provides you with moral purpose, why bother adopting a worshipful stance toward him? Also, if I think religions have socially pernicious effects, it is usually because they support a certain moral ethos. So, if I am interested in combating these pernicious effects, I will have to combat moral arguments.

Anyway, in light of my obsession, I am going to cover the section on moral arguments in Graham Oppy's book Arguing About Gods. This is not one of the stronger sections of the book, but he does dismiss ten moral arguments with reasonable aplomb.

I'll cover all ten, in ten separate posts. They should be fairly brief.

Here's the index:

1. The Argument from Objective Value

2. The Argument from Virtue

3. The Argument from Happiness

4. The Argument from Scripture

5. The Argument from the Need for Justice

6. The Argument from the Costs of Irreligion

7. The Argument from Heavenly Reward

8. The Argument from Conscience

9. The Argument from Convergence

10. The Argument from Practical Reason

Taxonomising Theistic Arguments (Part 1)

Tweet

This is a work in progress. Comments and feedback are welcomed.

Sabio, over at triangulations, recently did a post on taxonomies. Luke, over at commonsenseatheism, is trying to develop a taxonomy of worldviews. Biologists, apparently, have been at it for years. I want to add my voice to the growing chorus of taxonomy-fetishists by attempting a taxonomy of religious/theistic arguments.

In doing so, I am inspired by what I recently read in Graham Oppy's book Arguing About Gods. In the opening chapter, Oppy sets out ad-hoc taxonomy of theistic arguments. But he says:

I don't care how the arguments are grouped together; what really matters is that no arguments should be neglected.

Frankly, that kind of attitude baffles me. I love a well-developed system of classification, and I think it does make a great difference to how we approach and perceive the world. One of the more persuasive aspects of Darwin's theory of branching descent was how it fit the Linnaean taxonomic system.

Now, I have heard all the usual criticisms. It is said that taxonomies can be arbitrary; that they carve nature at its non-existent joints; that they turn a continuous reality into a discontinuous abstraction. The classic example of this comes from Jorge Luis Borges. His short story "The Analytical Language of John Wilkins", which is all about arbitrary classification, includes the following (fictional) extract from a Chinese encyclopaedia:

[I]t is stated that animals can be divided into the following classes: (a) belonging to the Emperor; (b) embalmed; (c) trained; (d) sucking pigs; (e) mermaids; (f) fabulous; (g) stray dogs; (h) included in this classification; (i) with the vigorous movements of madmen; (j) innumerable; (k) drawn with a very fine camel hair brush; (l) etcetera; (m) having just broken a large vase; (n) looking from a distance like flies.

I'm sure we can agree that this classification system is silly and paradoxical ("included in this classification" - wtf?). Therefore, all classification schemes are silly and paradoxical. Case closed, right?

No, of course not. The response to critics like Borges is not to abandon systems of classification, but to develop better ones. The reason is straightforward: they are all we have. I challenge anyone to come up with a way of viewing the world that does not rely on some sort of classification. I maintain it can't be done.

This is turning into a bit of a rant so I'd better get back on track.

Going back to Oppy for a moment, and to be fair to him, he does offer a pretty decent taxonomy of theistic arguments. He divides them into ontological, cosmological, teleological, minor evidential, moral and non-evidential (pragmatic e.g. Pascal's Wager). This is a fairly standard philosophical approach. I want to try something a little different.

As stated at the outset, this is a work in progress. I would like to see if people agree/disagree with my approach. Obviously, if you do agree or disagree, give reasons.

Anyway, I will to do two things in this post. First, I will consider the need for an organising principle for the taxonomy. Second, I will sketch my proposed taxonomy. To give the discussion some credibility and weight, I will use evolutionary "tree of life" taxonomies as a reference point.


1. Finding an Organising Principle: A Pilgrimage? 
All good taxonomies need an organising principle: some overarching cognitive goal that shapes and guides the process of classification. Bad taxonomies, like Borges's one, lack organising principles. To illustrate the point, consider the following (beautiful) diagram (you need to click on it to get the full effect):

This diagram is of the tree of life. The organising principle behind the tree is descent with modification. This is the idea that every organism is related by descent from a common ancestor. They look and behave differently because offspring can differ slightly from their parents (or parent!) and, given enough time, this process results in the huge diversity of life. The diversity and time is amply illustrated in the diagram.

The tree of life represents a journey through design space.* It starts with one parent organism at time T1, this parent has offspring at T2, they have offspring at T3 and so on. At each reproduction point, the offspring can spread through design space. What is meant by this is that although organisms vary along multiple dimensions (bone density, size of teeth, acuity of vision, sensitivity, brain mass etc.), and so occupy a vast swath of a multi-dimensional design space, they once started off in a localised, non-differentiated location (i.e. the last universal common ancestor or LUCA).

So to reiterate, the organising principle of the tree of life is "descent with modification". By fitting organisms (living and extinct) into the tree, we can represent life's journey through design space.

There is another important point to be made here: we can approach the tree in two directions. We could, of course, start at LUCA and work our way up through the madly divergent branches to the present day. Alternatively, we could, as Richard Dawkins does in his wonderful book The Ancestor's Tale, go on a pilgrimage back down the tree of life.

In other words, we  could start in the present day (at the tips of the branches) and wander back down to LUCA. As we do so, we rendezvous with our cousins (who start out on their own pilgrimage at the same time). Our first rendezvous is with our chimpanzee and bonobo brethren, our second is with the gorillas, and so on.

These two ideas -- the organising principle and the backwards pilgrimage -- are going to guide my approach to taxonomising theistic arguments.

2. A Journey Through Argument Space
The organising principle for my taxonomy of theistic arguments is going to be dialectical usefulness in journeying through argument-space. That's quite a mouthful, so let's unpack it and see what it means.

I am imagining groups of believers and non-believers having an argument over who has the right beliefs. Their argument takes place in a multi-dimensional argument-space. This "space" consists of different methods of argument (emotional, evidential, experiential, scientific, historical, and logical) and different types of argument (first cause, moral, design and so on).

 The goal of both sets of individuals is to convert the other set to their way of thinking. To achieve this goal, they cannot simply dive-in and use their favourite arguments. They have to figure out how best to sequence their arguments in order to convince the other side.

The best sequence of arguments, is the one that achieves the most common ground between the opposing sides. We will rendezvous on different common ground with different types of believers. If we can get the participants to use the same methods, and to focus on the same arguments, at the same time, we will achieve partial success. If we can get them to agree on conclusions, we will achieve total success.

This is what I mean by "dialectical usefulness in journeying through argument-space".

I hope you can begin to see how this is similar to Dawkins's backward pilgrimage: Dawkins envisaged all the currently existing species setting out on a journey to LUCA; I am envisaging all current believers and non-believers setting out on a journey to a common ground (yet to be defined).

There are two crucial differences between my journey and Dawkins's journey. First, Dawkins's journey has a temporal dimension: we are not just journeying through design space, we are journeying through time. This means that the branches on the tree of life can only ever grow in one direction: upwards towards the future, not downwards toward the past. I suggest we ditch this temporal dimension when considering argument-space. This means that the tree we ultimately draw will have offshoots and dead ends that need to be traversed.**

Second, in Dawkins's journey, we rendezvous with those we are most closely related to first and those we are least closely related to last. I suggest that the opposite is true in argument-space: to find the most useful path is to find a way to rendezvous with those most distant from your position first.

This may seem counterintuitive but I maintain it is correct because you might not know who you are arguing with. So, imagine three different people: A is a non-believing, scientific rationalist; B is a believing scientific rationalist; and C is a believing anti-scientific fideist. Now let's approach an argument from the perspective of A (I do this because it is my own perspective).

A does not know whether he is arguing with B or C. If he was arguing with B, things would be relatively easy: they would agree on methods for testing and confirming their respective beliefs, and they could focus on the available evidence and on particular arguments (cosmological perhaps). If he was arguing with C, things would be more difficult because they would not have an agreed-upon method for assessing their relative beliefs. It seems clear that if A wants to have a successful argument, he must assuming he is arguing with C. In other words, he needs to start by showing why emotional convictions are not a sound basis for beliefs and then try to develop a common epistemology.


Okay, that's enough for now. In the next part of the series, I will try to offer a more detailed elaboration of the taxonomy of arguments that follows from my organising principle. Before that, I would love to get feedback back from any of the (20?) people who read this blog. Do you think this approach is useful? What snags might it face?


*An abstract representation of the possible ways in which an organism can vary.


**The tree of life also has these, but as we journey back in time we only ever find the ancestors of species that went down these dead ends, we do not journey down them ourselves. In argument space we may need to go down these dead ends in order to rescue our fellow travelers.

Religious Experience (Part 1): Argument vs. Experience

Tweet

This post is part of my series on Nicholas Everitt's The Non-Existence of God. For an index, see here.

I am now working my way through Chapter 8 of Everitt's book, which deals with religious experience. In this first part I will consider whether experience is an alternative to argument or not. Some of this material overlaps with the previous discussion of Plantinga.


What are Religious Experiences supposed to do?

Religious experiences (REs) come in a variety of forms, but they all share some core features. In particular, they all purport to be experiences (perceptual or mystical) of God. Before getting down to the philosophical questions, consider the following example:

On a beautiful fall day, as I was hiking in the Cascade Mountain, the majesty and beauty of God's creation overwhelmed my resistance. As I rounded a corner and saw a beautiful and unexpected frozen waterfall hundreds of feet high, I knew the search was over. The next morning I knelt in the dewy grass as the sun rose and surrendered to Jesus Christ.

That is, of course, Francis Collins describing a religious experience he had. It was so powerful that it converted him to Christianity from atheism (see The Language of God - his experience may have be a "straw that breaks the camel's back" phenomenon).

What are we to make of Collins's RE? Two significant questions seem to arise: (a) If I had such an experience, should it convince me of the truth of God's existence? and (b) If someone else told me they had such an experience, should it be enough to convert me to their faith?

Defenders of arguments from RE (call them experientialists), have, on separate occasions, answered both  questions in the affirmative. We need to see whether they are right to do so.


Basic Beliefs Once More
The first line of defense for experientialists is to claim that REs give direct epistemic access to God. With such access, no argument is needed. This claim brings us back to the idea of a basic belief -- that is: a belief that needs no justification. Uncontroversial examples include my belief that "1+1=2" or my belief that "I am in pain". These beliefs are self-justificatory.

Are REs in the same category? Take my experience of the cup I am holding. I can see it, I can touch it and I can say that it exists. I do not need to infer its existence from my experience; its existence simply is the content of my experience. Experientialists claim that REs are like this.

Critics of these claims will argue that there are implied inferences taking place and that these inferences are open to challenge. It helps if we distinguish two varieties of inference:

• A-inferences: these are inferences that are used to arrive at a belief.
• J-inferences: these are inferences that are used to justify a belief.

The importance of the distinction is easily captured with the following example. Your romantic partner comes into the room, you look at them and say to yourself "He/she/it's angry". You arrive at this belief without making any inferences (no A-inferences). In other words, you do not look at them, examine their frown, identify a general principle stating that people with such an expression are usually angry, and conclude that he/she/it is angry. You jump straight to the belief.

However, if someone were to ask you how you knew that they were angry, you would be forced to justify your belief (use J-inferences). It would not be enough to say simply that your experience proved their anger.

The distinction between these two types of inference can be used to develop criteria for establishing whether something is a basic belief. If we consider an uncontroversial example, such as my belief that I am in pain, we see that it needs neither A-inferences or J-inferences: the experience of pain is the full content of the belief and if I told someone I was in pain they would not ask me to justify my belief.

So a belief that needs neither type of inference is definitely basic. What about beliefs that involve inferences of either kind? They could not really be included within the category of basic beliefs. Because they are based on inferences of some kind, they are always open to challenge.

Are Religious Experiences Properly Basic?

The final question to be addressed in this post is whether or not REs involve inferences. To answer this we need to look at some REs.

William Alston's 1991 book Perceiving God offered the following taxonomy of religious experiences (click to enlarge):

This suggests that there are two main varieties of RE: (a) experiences of God's properties; and (b) experiences of God's actions. Both of these types of experience involve inferences of some kind.

Let's go back and consider Collins's experience of the frozen waterfall. It seems clear that Collins's experience falls into the first of Alston's categories: through observation of nature, Collins experienced some property (glory or power) of God.

It seems equally clear that there must be an inference underlying the experience. Collins had a conscious experience of one kind (of nature), deemed it to be impossible unless there was a God, and so reached the conclusion that it really was a direct experience of God.

Indeed, all REs seem to involve the following type of inference:

1. I have had conscious experience X.
2. I would not have had experience X without there being a God behind it.
3. Therefore, my experience is of God.
4. Therefore, God must exist.

In sum, REs are more like "X is angry" than "I am in pain". And because they are like this they are open to all the challenges that arguments are usually open to. We will consider those challenges in future posts.

When Mechanistic Models Explain (Part 3): Evaluating Models

Tweet

This post is part of my series on the work of the philosopher Carl Craver. The series deals with the nature of neuroscientific explanations. For an index, see here.

I am currently looking at an article by Craver entitled "When Mechanistic Models Explain". Part one covered some distinctions between explanatory models and non-explanatory models. Part Two considered the example of the Hodgkin-Huxley model of the action potential and how it transitioned from being a how-possibly (non-explanatory) model to a how-actually explanatory model.

In this part, I will set-out Craver's evaluative criteria for mechanistic models. The criteria help to distinguish good explanatory models from bad non-explanatory ones. In deriving these criteria, we consider four questions:

1. Does the model cover all the relevant phenomena it purports to explain?
2. Does the model identify all the entities/parts involved in producing the target phenomenon?
3. Does the model identify the activities involved in producing the target phenomenon?
4. Does the model successfully organise the entities and activities?

Before looking at each of these questions in more detail, it is worth refreshing our memories about the nature of a scientific model. A model takes a target phenomenon (T) and represents it in some algorithm or function (S). S can then be implemented in a physical system, computer program, mathematical equation, box-and-arrow diagram or whatever.

(1) Does the model cover the target phenomenon?

The Hodgkin-Huxley model (S) is a mathematical model of the action potential (T). The action potential is the sudden change in membrane potential in a neuron. It is a multi-faceted phenomenon. It involves the sudden increase in potential, followed by a rapid decrease and recovery period. The first test of a successful model is whether it captures all facets of T.

The second test is whether the model can describe the inhibiting conditions of T. In other words, if we know that action potentials can be prevented or inhibited (e.g. by applying tetrodoxin, which blocks the flow of sodium ions through the ion channel), then our model should be able to account for these inhibiting conditions.

The third test is whether the model can identify the modulating conditions of T. In other words, the model should be able to tell us whether an alteration in certain variables will affect T. So in the case of the action potential, we want our model to tell us if variations in the density of ion channels or the diameter of the cell have an effect on it.

The fourth test is whether the model covers what happens in non-standard conditions. For instance, most laboratory experiments occur in non-standard conditions, the difference between these and standard conditions should be covered by the model.

The fifth test is trickier. One thing that distinguishes a how-possibly model from a how-actually model is that the latter will account for byproducts in the mechanism being modeled. So for example when sodium ion channels are opened there is a slight movement of positive charge to the outside of the cell, this is quickly counteracted by the flow of the positively-charged sodium ions into the cell. The slight movement of positive charge to the outside would not be an essential part of a how-possibly model of the action potential because it is an irrelevant blip. However, it would be an essential part of a how-actually model. More recent models of the action potential cover this blip. It turns out it is caused by a particular component of the ion-channel (the alpha helix, discussed here).

(2) Does the model identify the entities/parts responsible for the phenomenon?

Mechanistic models explain phenomena in terms of entities and activities. A key part to a successful mechanistic explanations is the correct identification of entities. By "correct" is meant actually existent not fictional.

The line between a fictional entity and one that actually exists is indistinct. For instance, one could argue that Mendelian genetics started out with fictional entities (genes), but that this definitely changed through repeated experimentation, manipulation and confirmation of Mendel's model. Despite the blurry lines, we can be confident that the entities actually exist if:

• They constitute a stable cluster of properties: for instance, ion channels were originally fictional, but over time it became possible to sequence their structure and work out how they reacted when in the presence of certain chemical agonists and antagonists. Thus ion channels were found to have a stable cluster of properties.
• They are robust: they can be identified with different techniques and devices. For instance, ion channels can be identified through pharmacological manipulations, X-ray crystallography, and electron microscopy.
• They can be used to intervene in other processes: for instance sodium ion channels can be manipulated so as to open potassium ion channels.
• They are plausible in the circumstances being investigated: this criterion varies from case-to-case. In the case of the action potential, any proposed entities should be plausible given background knowledge about the construction of nerve cells.
• They are relevant to the target phenomenon: in other words, when modeling the action potential we should not include entities that are present in the cell but clearly have no role in changing membrane potential. Examples might include: DNA, RNA, and microtubules.

(3) Does the model identify the relevant activities?

As mentioned above, a mechanistic explanation is built out of entities and activities. The activities are the things that the entities do (e.g. sodium ions diffuse down concentration gradients). It is possible to identify fallacious activities. For example, one could observe that the rooster always crows before the sun rises. But it would be a mistake to think that crowing was what brought about the sun rise. Mistaken causal inferences of this sort are everywhere (astrology, homeopathy etc.).

So how do we know if we have correctly identified the activities? Craver proposes one criterion that can help answer this question: the manipulability criterion. If we propose an activity linking two variables (e.g. roosters and sunrises) then it must be the case that manipulating one of these variables manipulates the other. Thus, silencing the rooster should -- if it were a correctly identified activity -- stop the sunrise. Of course it doesn't, so this is a bad mechanistic model.

(4) Does the model correctly organise the entities and activities?

The final crucial criterion for a good mechanistic model is its organisation. In other words, the model must have the correct spatial and temporal organisation of the entities and activities identified. It cannot be the case that taking out one entity and relocating it produces the same phenomenon.

There is quite a bit to take in here. In the interests of cognitive perspicuity, the following diagram summarises the criteria we need for evaluating mechanisms.