In February, I read the case of the ‘Inexplicable Omelette’, in which the author played off a difficult question: how can you tell whether something has come into its current form ― been constructed or altered ― with intent? His argument was that some things are too complex to consider unintended, and even an omelette is obviously intended. The language seemed off, and I’d like to explain how I propose navigating this question.
In discussion, the author threw around a lot of words with the implication they were mutually exclusive. ‘Intentional’ was one word, however its antonym was swapped around: ‘Unintentional’ was used, but so was ‘random’ and ‘chaos’, and these allow the author to muddy the water by tying complexity to intention. The first step is to explore this language use, as in rational discussion these words have places and functions. The second thing to do is create a defensible order to considering details to evaluate whether something is intentional or not.
The first thing to do, then, is to create a dichotomous pair: a word and its antonym that can be titles for categories that encompass everything, without overlap. The concept in question is ‘Intention’, so that is our first word. There’s no need for suspense: the other word is ‘Unintentional’. It is neither chaos nor randomness. It is worth discussing why not.
Randomness refers to the unpredictability of an event. The result of rolling of a fair die is random because the subsequent event is unpredictable. (The unpredictability can even be assigned a probability. This has its own interesting consequence: that many events, on average, do become predictable; that you can expect 4 to show up ten times in sixty throws.) That which is intentional and that which is random is not a complete set ― i.e. not everything is one of the two ― and neither are they mutually exclusive ― i.e. something can be both. Physics describes unintentional nonrandom behaviour of energy and matter: the trajectory of the moon around the Earth; the cooling of a freshly brewed coffee; the arc of a bouncing basketball. What is more damning for the author’s thesis is that this nonrandom behaviour can even have the property of creating order and complexity. Gravity creates stars; it makes matter gather and eventually coalesce. Entropy also facilitates complexity (see here and here). Why this is important ― that complexity doesn’t relate to intention ― is important later.
Chaos is about nonlinear systems, where small changes in initial conditions lead to big changes in the result. Where changes in the initial conditions matter at a scale more resolute than we perceive, chaos becomes identical to randomness. But this only happens at those scales. Chaos can be predictable (see this gif). Coin pusher arcade games are intentionally chaotic, the falling of a leave is chaotic without intent.
Our dichotomous pair must be ‘intentional’ and ‘unintentional’.
Now we have to device a method of evaluating whether something is intentional. To do this, I am employing Occam’s Razor ― not multiplying my entities unnecessarily. Intent itself is an entity, as is the agent that has the intent, so wherever possible one looks for explanations without intent. Failing that, an explanation using the intent of an agent we already know to exist is a lesser entity than speculating an agent we don’t know to exist e.g. droppings in the forest more better explained by a moose than Bigfoot. With that in mind, here’s the process:
(1) Do sufficient explanations exist, without calling upon intent?
If yes, end. If no, go to 2.
(2) Do sufficient explanations exist, calling on the intent of agents we already know to exist?
If yes, end. If no, go to 3.
(3) Could sufficient explanations exist, calling on the intent of agents we do not know to exist?
If yes, go to 4. If no, double check initial observation.
(4) Time to fund some research!
The best you have here is woolly conjecture and an hypothesis in need of investigation. Bear in mind, your research could uncover an intention-free explanation; it’s what normally happens.
(At point 3, the ‘Do’ becomes ‘Could’ to reflect the fact we are not talking about explanations that do exist, but instead on sheer imagination.)
Taking some examples the comments thread brought up, let’s look at a pile of sand compared to a three story sandcastle. Starting at stage 1, can we explain these things without calling on an agent with intent? Well, a pile of sand we can, sure: erosion. End. A sandcastle? Not so much, no. I mean, we could call on quantum fluctuations, but the probability is so vanishingly small that it isn’t a ‘sufficient explanation’. Go to 2. Can we call on a known agent with intent? Yes, people do build sandcastles. End.
(Note, I’m not saying this is a method of developing a complete explanation. I am saying this is a method of identifying intent.)
Now, as the initial discussion was simply a way of changing the words and sneaking the ‘design argument for God’, where this conversation goes next isn’t smooth; it’s a shoehorned talking point in the hope that I would lose track of where we were and have to admit to being wrong: is a human more like a pile of sand or a sandcastle? Admit it, you wanted to say ‘humans are more like a sandcastle’, didn’t you? The fact is the answer depends on your perspective and metric.
If you are talking from the perspective of complexity, then sure, humans are more like sandcastles. But we have already discussed whether complexity is at all relevant to the issue of intent. Remember, gravity and entropy are unintentional and yet facilitate, in a predictable fashion, complexity. But, as the implication is then than humans must require an explanation with intent, we see the metric and the perspective are being switched. From the perspective of explanations, humans are more like the pile of sand: a cumulative natural process over vast time, without intent, provides a sufficient explanation.
This bait and switch is now the fodder for any normal subject of the ‘design arguments for God’: the universe, life, complex life, consciousness and even intentions. Complexity and mystery are the crowbars to trick you into assuming intent. But, I think the structure I provided is defensible and helps you really see if proclaiming intent is justified.