Addressing human evolution || Part 1: The problem of human evolution

This post is part of a series on addressing the problem of human evolution. For a general introduction to the project, click here.

Steelmanning the argument: Why it's important

In this post, the broad question I'll attempt to answer is- why human evolution is a problem for Muslim apologetics. This would essentially consist of discussing all the putative evidence for evolution. I'm sure we've all heard of different lines of evidence for human evolution- the fossil record of hominids, high degree of similarity between humans and other creatures (especially chimps), so-called 'vestigial' organs and genetic elements in our body (like "junk" or non-functional DNA), etc. This post is not going to be a simple enumeration of this evidence. Such treatments are widely available on the internet, and a quick Google search will be more than enough for that purpose. What I do want to attempt here is to frame the evidence in a way that really brings out their evidential strength.

Let me spend a little more time to explain why such an exercise is important.

Popular, unsophisticated presentations of the evidence for human evolution often take the form of "talking points"- 99% DNA similarity between humans and chimps, what's the appendix doing in our body, why do our genomes share errors with other primates, and so on. When presented in this naive fashion, the savvy creationist can usually counter all of these objections point for point, without breaking much of a sweat either. However, I believe these popular presentations fail to communicate the true evidential strength of human evolution. As such, when expected to interact with these claims, the creationist interlocutor often misses the point in his responses. This misunderstanding of the evidence- the thrust- has led to the growth of a cottage industry of bad (or at least, inadequate) responses to evolution.

This is why I believe it's essential to first understand precisely what the evidential strength of human evolution is. Not only is this understanding a prerequisite to developing any good response to the problem (to avoid strawmanning), but also- and more importantly for my purposes- teasing out the thrust of the evidence in this way would clearly delineate what we must do to meet the burden of proof. A satisfactory response to human evolution must be built around the defined contours of its evidential challenge. As such, this post will be important in laying out the theoretical groundwork for the argument I eventually develop.

In much the same way that molds determine the shape of desserts, the evidence determines the shape a good response must take. Really in the mood for bad analogies today

So to that end, here's my specific game plan. I'll attempt to explain two concepts which I think are misunderstood (or under-emphasized) in discussions of the evidence for human evolution- one, the problem of gratuitous similarity, and two, evolution as an inference to the best explanation. In the course of clarifying each of these concepts, I'll also talk about some common responses to evolution, and show that they fail precisely because they fail to understand these two issues. Finally, I'll discuss the exact shape an adequate response to these problems must take, based on the foregoing discussion.

Two tiny disclaimers before we jump into this:

1. There's a population genetics-based argument against the creation hypothesis, which claims humans couldn't have originated from a single original couple. I'll discuss that argument as a part of this series, but it won't be relevant for our purposes today.

2. In this post, I use "evolution" and "human evolution" interchangeably- but know that I'm talking specifically about human evolution in each case.

The problem of gratuitous similarity

A lot of the evidence for evolution can be simply summarized as follows: humans are very similar to non-humans. The human genome is extremely similar to the chimp genome, certain aspects of the human anatomy is very similar to non-human ones, there are creatures in the fossil record that look very similar to humans, etc.

A lot of the time, popular presentations of the evidence can be reduced down to just that. Posed in that way, this objection isn't all too difficult to answer. Most cars have four wheels, so they are clearly similar in at least two ways: they have wheels, and they have four of them. This doesn't mean they evolved one from another- just that wheels are essential for them to function as cars, and four is a very pragmatic number of wheels for a car-like structure to have. So the similarity here can easily be explained on the hypothesis of shared functional constraints. Common design, one wants to say.

However, the similarity argument goes deeper than this. This is because the similarity between humans and non-humans has a peculiar feature: it is gratuitous.

Gratuitous means meaningless. Going back to the car example again- the four-wheel similarity between two cars is meaningful- they are similar because the shared features serve an essential purpose. They had to have been that way. What, then, does meaningless similarity look like?

Let me explain the concept with the aid of an analogy. I readily admit that the analogy is quite forced, but it's what I could come up with. I'll replace it as soon as I think of something better, I promise.

Take a look at the two cars below.

I stole the image off of Google and put the blue marks with MS Paint

The cars clearly seem to have a lot in common- wheels, mirrors, lights, seats, etc. However, there's a type of similarity between these two cars that are not like the others. There's a light blue mark- think of it as a strip of raised metal- at the back of both cars.

This similarity clearly doesn't serve a purpose. The blue strip isn't performing any function. In fact, this looks like a factory error- the metal being raised in that sort of unaesthetic, uneven fashion clearly gives off that impression. This type of similarity really is meaningless. Gratuitous.

A lot of the similar features between humans and non-humans can be characterized in just this way. We could've appealed to common design to explain shared features if they were required to perform the same function (there's a caveat though, read below). But we have a lot of features, especially in our DNA, that look like mistakes, errors, quirks, or kinks. Both human and chimp genomes are replete with corpses of viruses that once infected us, or shared mutations in otherwise functional genes, or bits of DNA that can move about willy-nilly but are otherwise non-functional. They are akin to that bit of off-color raised metal against an otherwise smooth surface. The fact that we share these features with other creatures in such a gratuitous way means it cannot be explained by citing shared functionality- common design- alone.

The creationist has a response to this, however. What if, he points out, these shared genetic elements aren't non-functional at all? What if they do have a function that we're either ignoring, or would be discovered in the future? Doesn't that solve the problem?

To illustrate the problem with this counter-response, let me modify the analogy a bit further- which will render it even more grotesque.

Let's say after close inspection of the car, we observe the following:

The weird blue rectangles are meant to represent drawers. Please just play along

Voilà. The blue raised bits of metal in the two cars that we thought were non-functional, actually seems to have a purpose- if you yank it hard enough, it opens up into a drawer at the back of the car where you can store stuff. So it does have a function- it's a drawer latch. Does this mean this similarity is no longer gratuitous?

No, of course not. Even if these two bits of metal function as drawer latches, this still doesn't answer everything about why they're similar. For starters, why are they both that particular shade of blue? Clearly, that bit of similarity- both of them being blue- doesn't have any functional significance. The drawer latches did not need to be such a specific color in order to function the way they do. So just because the shared feature itself has some function, as long as it has some properties that doesn't contribute to that function- they would still constitute an example of gratuitous similarity.

Let's export this example in the case of genetics and human evolution. The human genome has a high degree of similarity with the chimp genome, and large swathes of it aren't coding for proteins. However, it's plausible that all that extra genome is being used for something. For example, researchers (1, 2) have pointed out that the "bulk" of the genome is used to control the size of the nucleus. So while the sequence of the genome may not be very important, we still need the genome to be of a certain size in order to give the nucleus its appropriate size. This, in turn, is important for the cell. There's also some evidence that the genome needs to be "packed" in a certain way in the nucleus in order to have the specific functional roles that it does, and that packing requires the genome to be of a certain length. Meaning, all of that extra non-coding sequence in the genome is probably functional, only not in a sequence-specific way.

The genome is folded and arranged in particular compartments in the nucleus, and this arrangement has functional consequences. DNA here is working not as a code, but as structure. Picture taken from this paper

However, even if these shared parts of the genome are functional, that still doesn't explain why they have the specific sequence that they have. If their functionality is not sequence-dependent, then why do they have the same sequence? Compare with: if the functionality of the car drawer latch is not dependent on its color, why are they of that same peculiar color?

Todd Wood is a geneticist and a young-earth creationist who is generally very critical of bad creationist arguments. He responded to this "appeal to functionality" argument in a similar way on his blog:

A more subtle mistake is the insistence that vestigial organs have no function, that they are true evolutionary leftovers that are just waiting to be eliminated from our bodies. This leads to the equally fallacious response that by demonstrating a function - any function - the vestigial argument is nullified. In response to this reasoning, Steve Matheson made a great analogy to the function of a 1989 Yugo: Just because you can turn a Yugo into something useful, like a mailbox or port-a-potty, doesn't make it any less a vestigial car. In fact, it makes its car-ness all the stranger, since its automotive attributes have nothing to do with its function as a mailbox or whatnot. See my point? Having any function does not mean something can't be a vestige. I personally prefer to use the old term rudimentary structures for what we now call vestigial organs.

This underscores the problem of gratuitous similarity: It's not just that humans are very similar to chimps, it's that this similarity has no functional explanation- it could very well not have been.

And yet, here they are. And that's the problem.

In fact, this point can be strengthened even further. Even if one can show that a shared feature is functional in every way- that the drawer latch needs to be blue for some reason (let's say it's light activated, and only responds to a certain wavelength- look, I warned you this analogy would be weird)- this still doesn't solve the gratuitousness problem. This is because these ends could have been achieved in many other ways. If the function of the blue strip was to function as a drawer latch, couldn't the car have had a drawer or a storage unit in so many other ways? Why this way in specific in both cases?

This argument applies all the more to human design. There are so many different ways in which biological purposes can be achieved. The fact that the exact same features are being used in humans and non-humans to achieve the same ends- when that end could've been achieved differently (perhaps more efficiently)- still indicates gratuitous similarity. The biological anthropologist Jonathan Marks, in his book What it Means to be 98% Chimpanzee, sums up this point succinctly:

The 98% genetic correspondence of humans and chimpanzees does have a consequence with which hard-core creationists must wrestle— namely, that either humans and chimps do share a recent common ancestry, or else they have been independently zapped into existence by Someone lacking a great deal of imagination.

Basically, there are so many ways in which we could've been functional. The fact that we're functional in exactly the same way as chimps, is gratuitous.

This is what I call the problem of gratuitous similarity. The true evidential strength of this observation will become more apparent when considered together with the material below. However, the foregoing discussion has made the following point clear: a naive appeal to common design, even when joined with a claim of pervasive functionality of all of our features, cannot account for this problem. One would need additional argumentation to address this- a more nuanced creation hypothesis, perhaps.

This insight will be useful down the line when we develop our argument against human evolution.

Inference to the best explanation

We humans show significant similarity to a lot of different species- living ones, like chimpanzees and other primates, as well as extinct ones, like australopithecines and early members of Homo genus.

In addition to the aforementioned gratuitousness, there's another curious feature of these similarities relevant to this discussion. These similarities always form a pattern. And that pattern is best explained by evolution.

Consider the fossils. Australopithecines (left below) are anatomically less similar to humans than other hominids, and they show up first in the fossil record. Members of early homo (right below- Homo habilis) are somewhat more similar to us, and they show up next. Then comes the somewhat more "advanced" Homo ergaster, followed by Homo heidelbergensis or archaic Homo sapiens, followed finally by anatomically modern humans. The pattern that emerges from this is, creatures that are more similar to us show up later in history, while those more distant ones show up earlier. There's no real exception to this.

The point of this observation is: this pattern of similarity between humans and non-humans is very conducive to an evolutionary inference- simpler early, more advanced later. On evolution, this is exactly how we would expect the evidence to look.

Other observations about the fossil record only makes this conclusion stronger: for example, the most "primitive" technology is the earliest to appear in history, followed by a degree of advancement, followed by another degree still, and so on. So it's not like anatomical sophistication alone shows this pattern- technological advancement also broadly aligns with it. All of this, again, is exactly what you would expect on evolution.

Let me now substantiate the same point with the genome. Earlier I alluded to "virus corpses" shared between human and chimp genomes. This class of genetic elements- the endogenous retroviruses- are widely considered to be some of the strongest evidence for common descent. These elements enter the genome when viruses infect the cells and integrate their genomes into that of the host. As such, these are parasitic elements that usually confer no functional advantage. The fact that these elements are found to be inserted in both human and non-human genomes at the same locations is an example of gratuitous similarity par excellence. See this article for a brief description on this topic.

However, this similarity is not only gratuitous, but has a specific pattern. Look at this image that I took from an old paper (I first got wind of it from talkorigins): 

This paper came out in 2000, but serves our purpose well

This is a family tree of humans and the great apes. Each arrow signifies the existence of a specific type of endogenous retrovirus (ERV). These elements can be clearly seen to follow a specific pattern of distribution: the same elements are always shared between closely related species, and they never skip close family members for distant ones. You never see an element that's, for example, shared between humans and gorillas but not with chimps, who are closer to us. Never between chimps and orangutans, but not gorillas. So the pattern of distribution of these elements, themselves examples of gratuitous similarity, is very strongly indicative of evolution. Humans share a lot of ERVs with gorillas, but even more so with chimpanzees- because humans and chimps had a common ancestor and are part of the same family tree.

Todd Wood, whom I referenced above, makes the point forcefully in the context of the genome similarity between humans and chimps (reference):

[T]he mere fact of similarity is only a small part of the evolutionary argument. Far more important than the mere occurrence of similarity is the kind of similarity observed. Similarity is not random. Rather, it forms a detectable pattern with some groups of species more similar than others. As an example consider a 200,000 nucleotide region from human chromosome 1. When compared to the chimpanzee, the two species differ by as little as 1- 2%, but when compared to the mouse, the differences are much greater. Comparison to chicken reveals even greater differences. This is exactly the expected pattern of similarity that would result if humans and chimpanzees shared a recent common ancestor and mice and chickens were more distantly related. The question is not how similarity arose but why this particular pattern of similarity arose.

The upshot of this is the following. Humans share gratuitous similarity with other creatures in a pattern that's explained very naturally and comfortably on an evolutionary hypothesis. If evolution were true, these are exactly the patterns we would expect. In other words, these features are best explained on human evolution.

This distinction is important to understand. Many people, due mostly because of popular presentations of the evidence, think human evolution is a sort of deductive argument- the conclusion follows logically from the evidence. However, this is an unreasonable burden of proof for any hypothesis to bear. No theory needs to be preclude the logical possibility of an alternative explanation in order to be successful. Rather, human evolution is considered an inference to the best explanation- in terms of explanatory power and scope, simplicity, elegance and predictive capability- it's the best game there is.

Put differently, these lines of evidence don't logically preclude a creationist hypothesis from being true. God could've created humans separately, and yet for some reason, kept these markers of evolution in place. That hypothesis, together with the evidence, can both be maintained without logical contradiction. There are, however, two interrelated problems with this. First, these lines of evidence don't really "fit" naturally in a straightforward creationist hypothesis. After all, a creationist has no explanation for why these gratuitous similarities exist in the pattern that they do. Second, there's another hypothesis that explains the data much better. In scientific (and everyday) reasoning, an uncomfortable fit with the data, and the presence of a more compelling hypothesis, are two really strong indicators that the theory is probably false. Alternatively, evolution is true because it's the best explanation.

This is a basic point in epistemology that I expect my audience to be familiar with, so I won't belabor this point further. Let me round off this section with a quote from the University of Washington computational biologist Joshua Swamidass that I think really captures the intuitive punch of this problem (reference):

Let us imagine that God creates a fully grown tree today, and places it in a forest. A week later, a scientist and a theologian encounter this tree. The theologian believes that God is trustworthy and has clearly communicated to him that this tree was created just a week ago. The scientist bores a hole in the tree, and counts its rings. There are 100 rings, so he concludes that the tree is 100 years old. Who is right? In some senses, both the scientist and the theologian are right. God created a one week old tree (the true age) that looks 100 years old (the scientific age). Moreover, it would be absurd for the theologian to deny the 100 rings that the scientist uncovered, or to dispute the scientific age of the tree. Likewise, the scientist cannot really presume to disprove God.

Instead, the theologian should wonder why God would not leave clear, indisputable evidence that the tree is just a week old. My question to the theologians: Why might God choose not to leave evidence that this 100-year old tree is on week old? Alternatively, why might God choose to leave evidence that the week-old tree is 100 years old? [Emphasis in original]

Burden of proof

Putting everything together, the problem of human evolution is as follows.

1. There are numerous examples of gratuitous similarity between humans and other creatures. These examples are mined from different fields of scientific research, particularly comparative genomics and paleoanthropology.

2. These similarities cannot be explained on a creation hypothesis, except in an ad hoc way that adds unwarranted assumptions to the core hypothesis.

3. These similarities can be easily and naturally explained on human evolution.

4. (From 1, 2 and 3) These similarities are best explained on the human evolution.

5. Human evolution is probably true.

As mentioned at the beginning of this post- teasing out the main evidential thrust of the argument aids in conceptualizing possible solutions.

Things to come

If one takes the scientific evidence at face value- as I'll be doing, as mentioned in the last post- the only recourse for us is to attack (2). The burden of proof, then, is to provide a creation hypothesis that can explain this evolution-esque pattern of gratuitous similarity in a non-ad hoc way. Put differently, if one can produce a creation hypothesis that can easily and naturally explain this evidence, to the extent they follow from the standard predictions of the hypothesis- we can then say that evolution is not necessarily the best explanation of the facts, and, at least, there's a plausible contending hypothesis that can also be accepted without falling into irrationality. This is the minimal burden of proof that needs to be met in a good response to human evolution.

In this series, therefore, I'll be attempting to construct just such a hypothesis. To this end, I'll drawing on evidence from Islamic scripture and employ plausible theological reasoning. My main goal would be to show that this hypothesis is not ad hoc, and follows naturally from what we know about Allah and His actions (based, again, on scripture and theology). I'll then examine the specific lines of evidence for evolution once again (from paleoanthropology and comparative genomics), and account for them in light of this hypothesis.

While I believe this alone would achieve the minimal goal of providing a response to human evolution, I plan to do an additional, important thing as a part of this series. I will produce evidence that there are certain lines of evidence, just as significant (if not more) as the sort we've been discussing in this post, which cannot be explained on the human evolution hypothesis, but follow naturally from the creation hypothesis. If successful, this will leave us with the following two facts:

1. The "garden variety" evidence for evolution is explained just as well on the creation hypothesis (that I'll be defending),

2. Certain additional evidence can be explained easily and naturally on the creation hypothesis, but not on evolution.

The upshot of these two facts put together is, the creation hypothesis, and not the evolution one, is in fact the inference to the best explanation, just in terms of its explanatory scope.

All in all, my series would constitute a positive case of the creation hypothesis, and not just a defense, inshaAllah. In the next several posts, however, we will engage in a "fun and games" segment- where I critically analyze some extant unsuccessful attempts to address human evolution.

Series announcement: Addressing the problem of human evolution

Project introduction

In a series of posts starting from this one, I'll be attempting to lay out a robust solution to the problem of human evolution (i.e. the putative conflict between what Islam and science say about human origins). This first post is meant to define my claims and provide appropriate disclaimers, as well as outline the overall strategy I'll be pursuing.

I should note at the outset that this series is not meant to provide a comprehensive answer to every question related to evolution. The problem, by its nature, requires discussion of a lot of different disciplines- not just paleontology and genomics, but also population genetics, linguistics, philosophy of mind, neuroscience, and many others such besides. My readings in many of these fields is superficial at best. But in my opinion, tackling the most problematic brunt of the issue can be done without quibbling too much about a lot of the detail. For more clarity, of course, each argument I discuss in this series can be researched and developed to a greater degree of finesse.

The claims I'll be defending

Just this: Allah created humanity miraculously, in a way that cannot be explained on naturalism. The series will argue that this is the best explanation of all the relevant evidence (or at least: this is a better explanation of the evidence than naturalistic evolution). I will also defend the plausibility of humans emerging from a single original couple.

This is in stark contrast to other creatures on this earth, who, for the most part, came about through a process of natural evolution (so I'll be positing). Humanity, however, was a miraculous exception to this process, akin to how the clay birds made by Jesus (peace be upon him) were miraculous exceptions to regular birds. This claim that evolution is the "norm" and miraculous creation is the "exception", reserved for humans, will be important in defending the first claim above.

I think admitting that other creatures evolved is not theologically problematic at all. For a comprehensive defense of this, based on Qur'an, Sunnah and opinions of classical scholars, please see David Solomon Jalajel's book Islam and Biological Evolution.

This is one of the first books you should read about this issue

Note that I'm not claiming that Allah didn't supernaturally create (or otherwise intervene during the evolution of) any of the non-human creatures on earth. He could have, for example, supernaturally created the first life on earth, supernaturally caused the Cambrian explosion 530 million years ago that brought about all of animal phyla, or became directly involved with other parts of creation. I'm not ruling any of that out, and that's a separate investigation that fits better under the Intelligent Design research program. I am, however, claiming that for most of the history of life on earth, natural evolutionary processes have been operational. Perhaps somewhat simplistically: most of biological complexity on earth was brought about by natural processes (simplistic because, how do you measure what "most" of biological complexity is?).

Realism about scientific data

In attempting to provide a solution to the problem, I'll be taking the relevant scientific data at face value. So for example, I'll be accepting the reality, dating and interpretation of fossils (except of course, the part where I say humans didn't evolve from any of its fossil ancestors), and the verdict of comparative genomics about how similar we are to chimps. This is not to say there hasn't been any reasonable challenge to this data, and I plan to dedicate at least one post to what I think have been valiant, commendable attempts. The reason I choose to take the scientific evidence at face value is: most attempts at challenging the data of biology has either shown to have been false, or the objections don't really do what their proponents think they do (non sequitur). I'll be discussing some of these failed attempts in this series as well.

To use another popular topic in apologetics as an analogy: consider the age of 'Aisha (may Allah be pleased with her) issue. There has been two broad approaches to address it:

a) Accept that 'Aisha married the Prophet when she was nine, and then make sense of it on social, anthropological and moral grounds (this is a good example),

b) Argue, on the basis of scripture, that 'Aisha was married to the Prophet when she was much older (see this example).

My position about the scientific data on human origins is analogous to the first group's position about 'Aisha's age: I take the data at face value, and then attempt to make sense of it.

Outline of the project

So this is the general strategy I'd be pursuing.

1. Defining the problem

In the first post, I'll be laying out exactly why human evolution is a problem. This is just a statement of the putative evidence for human evolution, but I'll be framing this evidence in a way that I think really brings out why evolution is a problem that can't be ignored by Muslims. Specifically, I think most of the evidence for human evolution- fossils, genetics etc- can be summarized under what I call the problem of gratuitous similarity, and this set of phenomena is best explained on evolution. I think a lot of the confusion in addressing the problem stems from failing to understand the gratuitous aspect of the similarity between humans and non-humans, and the specific mode of reasoning used- inference to the best explanation. This will clearly delineate the burden of proof posed by human evolution, and help me frame my own solution down the line. There's also another argument which is not so much for human evolution as it is against Adamic origins: the claim that humans couldn't have originated from a single couple, based on population genetics considerations.

2. Unsuccessful solutions

To be honest, a large motivator for this project has been my constant frustration at seeing Muslims drop the ball on the human evolution issue.

As is standard practice in research, I'll begin by talking about a number of proposed solutions to the human evolution problem which, I think, all come up short in different ways. This is important because, as I'll show, there are lessons to be salvaged from these approaches, which would prefigure my own proposed solution. Also, engaging with the prevalent discourse would add to the comprehensiveness of this project.

Specifically, the approaches I'll be critiquing are:

a) Scripture can accommodate human evolution. The best case for this has been made by David Solomon Jalajel here.

b) Appealing to problems in the mechanisms in evolution (neo-Darwinism) to address human evolution. Many people consciously and unconsciously engage in this, including many proponents of the Intelligent Design movement.

Richard Owen was a contemporary of Darwin. He didn't buy Darwin's mechanism of how things evolved. That didn't stop him from accepting evolution

c) Appealing to scientific antirealism. This one, in my opinion, is probably the most grotesque, and yet somehow quite popular. Essentially, this involves undermining the epistemological basis for all of science to get back at human evolution. iERA's old essay on the topic is a good example of this approach, but I don't know if their methods have changed in recent years.

d) Addressing "talking points" of proponents of evolution, instead of the evidence itself. Just because evolutionist talking points are wrong, doesn't mean there's no evidence for human evolution (compare with this "climate denier" claim: "only" 93% of scientists believe in climate change, not the usually claimed 97%, so the problem is solved). Not to mention, a lot of the time- the criticisms of the talking points are themselves wrong. I'll be providing a lot of examples of this approach in my essay.

Of course, I won't be addressing the "meme"-esque refutations to human evolution based of blatant misunderstanding of the evidence (e.g. evolution is "just a theory", why are there still monkeys, etc).

3. Constructing an Islamic creation model

At this point, I'll take an important theological excursus to answer this general question: How does Allah create things? Or alternatively: Is there a general pattern to His action in the world? I think this is an extremely important question to answer, because that helps us address this other question: Assuming that the Islamic creation story is true- how would we expect the evidence to look?

It's absolutely essential we construct this model before we look at the actual evidence, because it's the tool which will help us interpret all of the evidence to follow. For this purpose, I'll first marshal some scriptural and theological considerations to construct and justify general expectations about Allah's creation. I'll then list a number of predictions which follow from this. The rest of this series would then just be holding up individual pieces of evidence, and assessing whether it confirms our predictions or not. My strategy would be to show that the evidence confirms all of the predictions of the Islamic creation model. On balance, then, Islam would be a better explanation of the human origin evidence than evolution.

I think this section is the most original aspect of this project. That said, this prediction-first-evidence-later approach was used by Fazale Rana and Hugh Ross in their book Who was Adam? A Creation Model Approach to the Origin of Humanity, even though I think some of their predictions were poorly thought out, contradicted most of the evidence, and almost all of their major conclusions were wrong. This goes to my earlier point about there being salvageable lessons even in an ultimately unsuccessful approach.

Funny story about this book: it first came out in 2005, and then in 2015- the authors added a new section to the book to account for all of the new data that had become available. As I said, the conclusions of the book are wrong, but it's very useful to read- especially when you're just starting to study this problem

4. Examining the evidence: fossils and genetics

This step onward would be looking at the putative fossil and genetic evidence for human evolution. To evaluate the fossil evidence, we would actually have to make educated guesses about when Adam might have emerged, and which fossil hominids can be considered his children, and which, in all likelihood, can't. On the genetics side, I'll be looking at different vestigial genetic elements like endogenous retroviruses and other mobile elements shared between humans and chimps, the putative chromosome fusion event, and the high degree of similarity between our genomes that can't be explained by functional constraints. I believe the responses I develop for vestigial genetic elements would apply to vestigial parts of our anatomy as well.

5. Examining the evidence: language and consciousness

Humans are so much more than just bones and genes. In order to evaluate predictions about human origins, we can't limit ourselves to just those two, but also consider other things that make us human, like symbolic thought, autobiographical memory, self awareness, and other such features. Here again the approach remains the same: assessing whether the existence of these features follow from predictions of the Islamic creation model.

This way of "expanding the evidence base" is essential to my approach, and is reminiscent of this essay by Yasir Qadhi and Nazir Khan. Of everything I've read from Muslims on evolution, this is probably the article I have the fewest number of issues with.

6. Population genetics issues

As alluded to above, there are some arguments based on population genetics which claim that humans could not have arisen from just an original couple. I'll be addressing this argument, mostly based on the work by Joshua Swamidass, Ann Gauger and Ola Hössjer, and Richard Buggs.

7. Alternative facts and loose ends

In the last parts of this series, I'll be discussing two things. First, as I said, I'll be mostly taking the scientific data on human origins at face value. However, there are a handful of researchers who have brought about some interesting reasons to doubt parts of this data. I'll be discussing some of this as promising research avenues to pursue. I'll probably also tie up any other loose ends in my argument at this point.

I'll end the series by talking about the future of all this. After all is said and done, I think (as the series would hopefully demonstrate) human evolution is a tractable problem in apologetics. Unlike Christianity and Judaism, when it comes to human origins, Islam has a very modest burden that it expects us to defend. I think that burden is altogether reasonable, and supported by the weight of the evidence. As such, I believe this problem can be solved conclusively within this generation, and we can move on to other things.

Parapsychology and Muslim apologetics

Definition of parapsychology

Courtesy of Wikipedia:

Parapsychology is a field of study concerned with the investigation of paranormal and psychic phenomena which include telepathy, precognition, clairvoyance, psychokinesis, near-death experiences, reincarnation, apparitional experiences, and other paranormal claims. It is often identified as pseudoscience.

I've read a couple of parapsychology books in my time- both of them by the analytic philosopher Stephen Braude. The Gold Leaf Lady and Other Parapsychological Investigations is a great gentle (but analytically robust) introduction to the topic, while The Limits of Influence: Psychokinesis and the Philosophy of Science is more rigorous and data-driven.

In this article, I want to make the case that parapsychology is an important avenue that Muslim apologists need to look into. To that end, let me provide a brief sketch of the ways in which parapsychology interacts with apologetics.

Debunking Naturalism

The worldview of naturalism- a commitment to a metaphysics and ontology that only involves things reducible to physical "stuff" i.e. entities and laws postulated by physics- continues to be the dominant strain of thought in academia. I think it's weird that naturalism has been able to persist for so long. Depending on which version of naturalism you adopt, it goes against the most fundamental of human intuitions- teleology in nature, soul/consciousness or agent causation in general, libertarian free will, human exceptionalism, the objectivity of ethical and aesthetic values, probably even the first person perspective. Philosophers have teased out all sorts of hidden consequences of naturalism- Victor Reppert, for example, argues that eliminative materialism entails that beliefs don't exist. Naturalists, of course, are aware of this. For them, the triumph of naturalism implies the defeat of folk, "superstitious" beliefs handed to humans as spandrels of biological evolution. So the fact that it grates against commonly held intuitions is seen as a natural prediction of the worldview.

Philosophers of a certain persuasion have been attempting to rebut naturalism for quite a while. While I think their case, especially when taken cumulatively, is very strong, parapsychology is an interesting player in this game. If paranormal phenomena can be shown to be veridical, then naturalism would be false as a matter of empirical fact. This would be a brutal and decisive rebuttal of naturalism, and would raise the prior probability of Islam in specific and theism in general.

There are two principal ways in which parapsychological phenomena are usually verified. First is via critical analysis of testimonies. If a parapsychological event- levitation or psychokinesis, say- occurs in the presence of a group of experts, and they subject the event to sufficient analytic testing to rule out foul play or natural explanations, and the witnesses pass other criteria of reliable testimony- then that event probably happened. This basically constitutes a commonsense way of thinking about testimonial evidence.

The Gold Leaf Lady is the first book I read on parapsychology. Interestingly, most of the stories in the book are about frauds that got caught, and other such mishaps

This bland, simplistic description of course invites a lot of skepticism. While I'm sympathetic to that feeling, I think a good way- the only way?- for someone to be convinced of the feasibility of such a research project is to either read about the stringent controls investigators often use to verify these phenomena (read either of the two books by Braude that I referenced above), and/or the sophisticated epistemological criteria developed by analytic philosophers of religion to sift out the genuine from the bogus (read Kai-Man Kwan's essay in The Blackwell Companion to Natural Theology).

As a sample of how rigorous the controls can be in a parapsychology investigation, here's part of a seance report from 1908 (cited in Braude's The Gold Leaf Lady, p. 49-50]:

[Medium’s feet tied to rungs of chairs of controllers on each side of her, length of rope on left being 20 ins., on right 21½ ins. Hands tied to one another—distance apart 22 ins., also left hand tied to B.’s right, distance being 16½ ins.; medium’s right hand tied to C.’s left, distance 18 ins.] 

10.12 p.m. Séance begins.Light I [brightest light]
Tilts begin almost immediately.F. Table tilts on the legs away from her.
10.13 p.m. Complete levitation of the table.C. Her right hand resting on the table touching mine, my wrist being between hers and the edge of the table. Her right foot in contact with my right foot. I saw a clear space of about eight inches between her dress and the leg of the table.
B. Complete levitation of the table for a second time [during dictation of B.’s control. Dec. 14/08].
B. My right hand on both her knees. Her left foot touching my right foot.
10.14 p.m. Complete levitation for a third time.B. Another complete levitation.
F. Both medium’s hands completely on the top of the table touching C.’s and B.’s.
C. I can see a clear space of about 8 inches between her dress and the table leg all the way down.
B. My right hand on her two knees. My right foot against her left foot, and I can see between her left leg and the table leg.
10.16 p.m. Another complete levitation.F. Her right hand off the table altogether, left hand on B.’s, pulling it upward and the table appeared to stick to it.
... 

[The light was sufficient to read small print by with comfort, at the further end of the room, the hands were always plainly visible and always situated so that it was clear that the table was not lifted by them. The extreme rapidity of the levitations made complete descriptions almost impossible, and it was decided to confine the description of the control to the feet, the control of the hands being obvious to all and description rendered unnecessary. . . . F., Dec. 14/08] (Emphases in original)

These reports are detailed and robust enough to count as reliable testimony, and there's no non-ad hoc way of rejecting them. Put differently, and this is the case epistemologists often make in defense of parapsychological phenomena, many of the best examples of these events are attested with a level of rigor equal to, or even more than, our usual standards of evidence.

Arrangement of the 1908 seances- the medium was Eusapia Palladino

Another way these phenomena are verified is by standard anthropology. If reports of a certain class of parapsychological phenomena are consistently associated with some specific details- even when the reports are collected across different cultures and time points, from people who had no contact with each other- then that probably speaks to a veridical core of these phenomena. The strength of this case depends on how widely shared the details are, and how peculiar and specific they are. Braude gives the following example in Limits: across many different cultures and time, poltergeist phenomena are always associated with

...the slow and gentle trajectories of airborne objects, the apparent passage of levitated objects through walls and closed doors, and the poltergeist bombardment with human excrement.

These details are way too specific to be just coincidences, much less resulting from "shared human experiences". Similarly, research on near death experiences suggest their widespread association with a wide range of peculiar details. Perhaps the most dramatic examples of these cases are when you have commonalities between modern and ancient or primitive reports. Andrew Lang in his The Making of Religion (review here) points out that the modern practice of gazing into a crystal ball to predict the future has its analogs in many "primitive" cultures, in the general form of "gazing into a smooth deep"- be it a drop of ink, or blood, or the surface of water. Unless there's a worldwide conspiracy involving millions of individuals spanning thousands of years (there are striking similarities between details of certain parapyschological phenomena reported by the ancient Greeks, and those documented in modern times), the only explanation of these shared features is that people are experiencing and reporting the same, genuine events.

Negative apologetics

Many parapsychological phenomena seem to, ostensibly at least, run counter to Islam's truth claims. Examples include:

a) Curing of possession/healing from diseases when names of other deities are mentioned;

b) Putative evidences for reincarnation in the forms of verifiable past life memories, biological signs corresponding to manner of death in previous life, acquiring proficiency in a different language, and so on;

c) Astrologers accurately and precisely predicting the future, clairvoyance;

d) Other 'miracles' happening associated with other religions or deities, and so on.

Our scholarship, and even some hadith reports, recognized some of this- the explanation is usually given in terms of the activity of the jinn (see the last section of my previous article). In order for that explanation to be a serious contender in parapsychology research, as opposed to just an ad hoc theory-saving attempt in favor of Islam, it needs to be made more precise with both scriptural and "field" data. Specifically, the explanations need to focus on patterns and regularities in jinn activity, association of these activities with specific parapsychological phenomena, and so on. In other words, jinn-based explanations should be of the following form:

P1. Phenomenon X exists.

P2. Jinn are known to cause X.

C1. Jinn are a plausible explanation of X.

The information in P2 would be known from scripture, "field research", and even anthropology (see last post in this blog). To see the importance of this, consider how this argument would have to be run if that information was missing:

P4. Phenomenon X exists.

P5. Possibly, Jinn can cause X.

C2. Jinn are a possible explanation of X.

P5 is almost impossible to dispute, because jinn occupy the realm of the supernatural and its hard to set limits on what they cannot do. However, just because it's possible that jinn caused X, doesn't mean jinn activity constitute a good explanation for X. The second formulation of the argument is too ambiguous to be falsifiable- unlike the first formulation, in which case one could set limits on what jinn are and are not known to cause. So in order to provide good, falsifiable explanations for these parapsychological phenomena with problematic theological content, we'd need to know more about the jinn- hence parapsychology.

Incidentally, I should also mention that there are alternative, epistemologically-based explanations for such "problematic" parapsychological phenomena as well. In attempting to explain near death experiences that have content contrary to Islamic teachings, one hypothesis one could pursue is - perhaps people have a warped access to the "world" where these experiences come from. So while there's a veridical core to near death experiences, other details of these experiences could've been the result of the worldly, cultural experiential filter they pass through. Perhaps similar explanations could be used in the case of other parapsychological experiences as well.

Miscellaneous

Two other ways parapsychology interacts with the case for Islam are:

a) My friend Sharif's research demonstrates that Islam's descriptions of the spiritual realm very often aligns closely with people's experiences of the paranormal. In a recent conversation, for example, he mentioned that people often report the existence of a guardian angel, a constant companion (qareen?), record-keepers responsible for each individual, etc. I don't know what the state of this research is, but it's very interesting that Islam accurately predicts details of the spiritual realm (under the reasonable assumption that there's a veridical core to people's religious experiences). This article also seems to be a good exercise along these lines.

b) Parapsychology also complicates the case for Islam, at least theoretically. The case for Islam is usually presented in the form of a miracle: Islam/Qur'an/the Prophet's life has feature X, feature X is a violation of the laws of nature, hence Islam/Qur'an/the Prophet is from Allah. However, as scripture itself attests (and parapsychology confirms), there are non-divine sources of "miracles" as well. This means one would need to make theoretical distinctions between divine and non-divine miracles. Our scholars have discussed some of these- e.g. it cannot be learned as a skill, it cannot be replicated, it must come with the proper context of a message, and so on. While not a huge problem, I do think some groundbreaking is required here, and these theoretical considerations need to be developed further.

Final musings

As this very, very broad strokes survey attempts to demonstrate, parapsychology is indeed a field where there is a lot of potential and promise for apologetics. And that concern brings us to a sad fact. As I've been recently realizing, to contribute significantly to apologetics, one should ideally make it their career. It's difficult to pursue apologetics as a "side hustle", e.g. become a scientist as a profession and study women's rights at your leisure. You can only contribute so much that way, and "so much" is definitely not enough given the amount of work that's left. Now certain other fields like biology, cosmology, history, and even sociology are more promising in this aspect. Researchers in these fields can make a living while pursuing these apologetics-conducive careers. The same cannot be said for parapsychology. It's highly unlikely that someone can make a respectful living out of a ghostbuster career. That only underscores the amount of funding and such we require if we want to see the success of Muslim apologetics through to the end, inshaAllah.

Bones and dung as sustenance: How anthropology of religion can help us understand the Jinn

The problem

In the prophetic ahadith, we find some peculiar details about feeding habits of the jinn:

The delegate of Jinns of (the city of) Nasibin came to me--and how nice those Jinns were--and asked me for the remains of the human food. I invoked Allah for them that they would never pass by a bone or animal dung but find food on them. [Source] 

They (the Jinn) asked him (the Holy Prophet) about their provision and he said: Every bone on which the name of Allah is recited is your provision. The time it will fall in your hand it would be covered with flesh, and the dung of (the camels) is fodder for your animals. [Source]

These and other ahadith suggest that the jinn feed on material things. But if that's the case, why do the material things- the bones and dung in this case- not simply disappear?

This should be an easy problem to solve. Jinn are supernatural entities, so there's no reason to expect that their feeding habit would be similar to ours. Maybe they don't eat the bones and dung themselves, just the supernatural "essence" of it. The second hadith actually makes this more explicit- when the bones are accessed by the jinn, flesh appears on them supernaturally, and that's what they eat.

My goal in this post is not so much to solve any problem, but to use this example to highlight a point: studying anthropology of "primitive" religion can either corroborate, or at least help us make sense of, some descriptions and activity of the jinn.

I want to start with some anthropology of religion, except framed from an Islamic demonology perspective. Bear with me.

Primitive animistic deities were probably jinn

Animism is a form of religion centered around "nature spirits"- belief in, and some sort of worship of, spirits living in trees and rocks and other inanimate objects. I think a strong case can be made that at least some of these "spirits" are jinn, and this form of religion was inspired by jinn-centered phenomena. For one, Andrew Lang makes a strong case that at least some primitive religious beliefs were inspired by genuine supernatural phenomena. His case is based on the eerie commonalities between primitive spiritual beliefs as reported in anthropology, and modern day parapsychological phenomena. You can read my review of his book here for a brief digest of his case. More generally, however, this also goes to show that existence of jinn and their behavior offer a ready explanation of how pervasive animistic beliefs are among humans.

Andrew Lang and his 1898 book, where he makes the case that a lot of primitive religious beliefs were inspired by veridical supernatural phenomena

Secondly, and perhaps more acutely, the behavior of the spirits as documented by anthropologists is very reminiscent of Islamic descriptions of ill-intentioned jinn. In the words of Winfried Corduan, the spirits tend to be "fickle, demanding and selfish" (reference below). The typical mode of "worship" offered to these creatures is not sincere, prayerful devotion- but mostly material offerings, as a negotiating tactic to extract favors. These behaviors explain why animistic religions tend to be unmoral- they don't come associated with a moral code for people to follow, simply because the deities and their commands are too transactional to base a moral system on. In sum, supernatural beings constantly demanding offerings from people in exchange for materialistic favors, and not inspiring piety of any sort- this description seems to fit well with how Islam portrays ill-intentioned jinn.

Finally, from an Islamic perspective, its very plausible that the jinn are responsible for polytheistic beliefs. Indeed, a hadith tells us how shirk began on earth- Satan, himself a jinn, inspired people to worship statues [Source].

Taking all of this together, it seems quite plausible that at least some of the primitive animistic deities were, in fact, jinn.

How did these jinn-deities eat?

Once we accept that fact, an important conclusion follows: the behavior of these spirits in animism could tell us about the behavior of jinn. In that connection, let's now look at their feeding habits to see if they shed light on our opening ahadith.

In his book In the Beginning God: A Fresh Look at the Case for Original Monotheism, Winfried Corduan discusses how "offerings" to these spirits would look:

Animistic cultures typically make abundant sacrifices to the spirits; frequently, they speak of “the spirits being fed.” If I may take an example from a highly developed culture, it is an important part of Chinese religion to present food to the ancestor spirits. The spirits, if they are so inclined, will consume the “essence” of this food and leave behind its material remnant, which can then be eaten by human beings. Sacrifices, in this sense, are ways of negotiating with the gods and spirits. The spiritual beings profit from what is being given them, and the human being hopes that his sacrifices will generate particular help from the spirits or gods. This attitude underlies much of the motivation in animism. [p. 143 in EPUB version. Emphases mine]

Some interesting themes to emerge from this are:

1. The spirits demand and feed on material offerings, not supernatural ones;

2. Their mode of feeding is supernatural- they don't chow down on the food itself, just absorb its "essence" somehow, leaving the material behind.

As it happens, these descriptions apply to the ahadith's description of jinn feeding habits as well: the stuff they eat from is material, but the mode of eating is supernatural so the "food" doesn't disappear. In this case, granted some plausible premises, anthropology of religion offers confirmation of the hadith descriptions of jinn feeding.

Anthropology of religion and demonology

I think anthropology of religion, especially "primitive" religion, is an extremely useful discipline for apologists to look into. Islam makes some specific claims about world religions (e.g. the first humans were monotheists, for example), and work done by anthropologists like Wilhelm Schmidt is definitely relevant in that regard. But there's another important utility. As this post was meant to demonstrate on a small scale, the objects of worship in many of these "primitive" religions are jinn, and anthropological literature about beliefs pertaining to these beings would be useful from a demonology perspective. This information could either corroborate what we already know from scripture- like in the feeding case- or, perhaps more interestingly, they can offer clues about their behavior that scripture is silent on.

This is useful, because we often reach for demonological explanations to account of "non-Islamic" supernatural phenomena- like exorcism in the name of deities besides Allah, certain forms of clairvoyance and fortune-telling, and perhaps most importantly, memory and behavior suggestive of reincarnation. These are supernatural phenomena that, in their theological content, are contrary to Islam. The practice of explaining them in terms of jinn is something that scripture itself endorses. For example, a hadith in Ibn Majah tells us that the wife of Ibn Mas'ud (may Allah be pleased with him) used a particular amulet that prevented her eyes from watering. Since amulets, talismans or charms are haram, this would be an example of a non-Islamic "faith healing". Ibn Mas'ud explained this phenomenon by appealing to ill-intentioned jinn- a jinn was causing the eye problem to begin with, and he stopped it when, as a sign of obedience to him, the amulet was worn [Source]. Another example would be the Prophet's hadith on soothsayers: when some people told him that, contrary to Islamic beliefs, the soothsayers seem to know some details about the future somehow, the Prophet invoked the jinn as the supernatural source of this information [Source].

This book is a great example of how demonology can provide theologically sensible explanations of supernatural phenomena- UFOs in this case

As such, to explain religious experiences with problematic theological content, we often need to use jinn as explanations. An attractive example here is the so-called "reincarnation evidence". There are some instances where a person can recall accurate, verifiable details about the life of another, now deceased, person, even though there was no way of knowing this information. In some cases, the person can even speak in the accent of the deceased, display the mannerisms peculiar to them, or even speak a language they didn't know before. While these could be suggestive of reincarnation, Islamic demonology also provides a specific way of explaining this set of phenomena. We believe each human individual is constantly accompanied by a jinn [evidence]- an entity that knows everything there is to know about us (Wikipedia sums it up nicely by calling them our "spiritual double"). So in case of "past life recall", a jinn-explanation would be- the spiritual double of the deceased person has possessed the recaller. This of course would also explain sightings of the deceased, as jinn can shape-shift.

In order to provide explanations of this sort, and for these explanations to be good, we need more information on jinn behavior and activity. Sources of this information consist of scriptural data, and "field research" by Muslim exorcists (raqis). However, given the theoretical and practical considerations laid out in this post, anthropology of primitive religion could also be another important source of information in this regard. In the same vein, it should also be useful to study jinn-related religious anthropology in Arabia.

I think the utility of this project is better appreciated by someone who understands the challenges posed by problematic parapsychological phenomena in general. My next article would be on that broader topic.

Why did Allah assign angels to record our deeds, when He could've done so Himself?

The problem

The Qur'an tells us:

"And indeed, [appointed] over you are keepers, noble and recording- they know whatever you do." [Sura Al-'Infitar, 82:10-12]

A commonly heard concern about this is, why did Allah need to appoint angels to record our deeds? Clearly, as the Qur'an repeatedly emphasizes, He has intimate knowledge of our every thought and action. So isn't it redundant, then, to post additional record-keepers on our shoulders?

I think this is an interesting discussion, because it serves to flesh out some principles about Allah's actions.

The basic response is as follows.

Allah as a "motivational creator"

A lot of the times, Allah creates things not in the way that's the most "efficient" in terms of resource, but in a way that would serve to make a spiritual point. A good example of this is how He chose to create humans. From a purely "efficiency" perspective, creating Adam de novo from dust, and then making all of humanity from the initial couple seems more trouble than its worth. If Allah wanted, He could definitely have had us naturally evolve from apes. The reason He chose to create us supernaturally, however, was so His Prophet (peace be upon him) could make this point after the conquest of Makka:

“O you people! Verily Allah has removed the slogans of Jahiliyyah from you, and its reverence of its forefathers. So, now there are two types of men: A man who is righteous, has Taqwa and honorable before Allah, and a wicked man, who is miserable and insignificant to Allah. People are children of Adam and Allah created Adam from the dust.” [Tirmidhi, English reference- Volume 5, Book 44, Hadith 3270]

In other words, this mode of creation allows Allah to make a moral point to people: don't be racists. Racism, on Adamic creation, is not only immoral, but also unscientific: contrary to reality, since our stock is the same. Also see Qur'an, 49:13.

Similarly, in many other cases, Allah sets things up so He can make arguments on their basis. Adam was created from dust, and not from light, not because dust is easier to create from- but because it serves to underscore our lowly status as creation. We as humans can draw significant psychological motivation from that fact. In the same way, Allah brought His Prophet up to Him on the night of ascent to reveal the commands of prayer. Again, this seems superfluous- He already had a streamlined Jibreel-based system set up to deliver revelations to the Prophet- but the action makes perfect sense when we realize this was done to emphasize the gravity and importance of prayer.

In the same way, while awareness of Allah should be motivation enough for us to do good deeds and stay away from bad deeds, the idea that we have angels literally atop our shoulders, breathing down our necks as it were, busily taking note of every single action we commit and reporting to Allah- that serves as far greater psychological motivation. That's the reason the angels are posted, not because this is how Allah "comes to know" about our actions- but because this setup serves as additional motivation for us to watch what we are doing.

I kind of want to substantiate this by talking about a real life example about the specific ways religion motivates us.

So let's talk about Balinese water temples.

The rice terraces in Bali



Pictured above are rice terraces in Bali. All of the quotes in this section are from David Sloan Wilson's book Darwin's Cathedral: Evolution, Religion, and the Nature of Society, pp. 132-138 in the EPUB version. He bases his analysis on the work done by the anthropologist J. Stephen Lapsing, in his book Priests and Programmers: Technologies of Power in the Engineered Landscape of Bali.

Apparently rice is an exceptionally difficult crop to grow, especially in terms of water management- paddies need to be efficiently irrigated and drained at precise times in the planting cycle. Describing the overall structure of the rice terraces, Wilson writes:

The rain that falls on the mountain tumbles down to the sea in rivers that cut deeply into the soft volcanic rock. To use the water for irrigation, the Balinese have created a vast system of aqueducts, often running through tunnels a kilometer or more in length, that shunt the water from the rivers to the rice terraces, also sculpted by human labor from the steep mountain slopes.

Balinese rice farmers have fine-tuned this whole process to maximum efficiency. To understand what a marvel of pre-industrial engineering this is, consider these three facts: "First, the Dutch, who conquered Bali in the nineteenth century and whose own country is a miracle of hydraulic engineering, could find little to improve in Balinese rice production other than to tax it." Second, recent attempts at "modernizing" this system with pesticides, fertilizers or other interventions have been runaway failures. Third, and perhaps most impressively, computer simulation models of the system showed that it was "close to optimal at solving the trade-off between water use and pest control".

Here's the thing, though. This system really shouldn't work, at least not as well as it does.

"The goddess makes the waters flow..."

This is because maintenance of a system such as this requires collaboration among thousands of people. For the water to be distributed efficiently and equitably, a farmer upstream the irrigation branch cannot misuse the water- that would deprive those down the river from him. At each branch of the irrigation system, fates of several groups of farmers are joined by their shared use of the water. The functionality of the system, therefore, depends massively on honest collaboration among all those who use the water.

I'm massively underplaying how complex this problem is, because water management is just a part of the problem. There's also pest control, which also requires large-scale collaboration among hundreds if not thousands of individuals.

Without a strong central government or policing system, there's no reason this should work. And yet, Balinese rice farmers don't have a system of that scale. How does this system operate as smoothly as it does?

The answer, in one word, is religion. At the summit of the mountain, there's an immense temple dedicated to the worship of the goddess of the waters. The farmers' "overflowing belief" in the authority of this goddess, and her high priest, is what leads them to maintain the rules of charity and collaboration as well as they do.

Now this in itself is an interesting story about the secular utility of religion, but there's one more layer to it. The Balinese irrigation system depends not only on just belief in the water goddess itself, but also plenty of other, smaller temple deities that are placed at every branch. The farmers fed by the water from a branch form the congregation of that temple. So it's not only the more "distant" water goddess that the farmers are held accountable to, but also to the smaller-scale temples (and deities) immediately upstream of them.

The general idea, to the extent I can place it, seems to be that with deities "watching over them" right at the point where they receive the water, the farmers are even more motivated to cooperate.

In other words, the fact that the water deities are proximate, not distant- the fact they're placed in charge of upstream irrigation branches- serves as important motivation for the rice farmers. Circling back to our initial discussion: accountability to a proximate spiritual agent- angels on our shoulders, in this case- does lead to a greater psychological motivation to do good.

That's essentially why Allah set things up the way they are.

Final thoughts

An interesting point to emerge from this discussion is, angels in Islam seem to be doing the same explanatory work that minor (in some cases major?) deities perform in polytheism. I've heard some arguments about polytheism providing a better explanation for some aspects of reality. This tiny demonstration shows that, while Islam is an extremely strict monotheistic religion, the concept of angels do provide explanatory resources for phenomena thought to fit better with polytheism (like the dappled nature of the world, say). What's different, then, between (at least some forms of) polytheism and Islam is not the belief in a multiplicity of powerful, honorable spiritual beings, perhaps not even in the roles they serve in their respective theologies- but the response that's expected of us towards these "lesser" beings (honor and respect, not worship).

These strands can perhaps be fleshed out into a robust yet spiritually engaging Islamic Angelology.