Can we ever infer design purely from improbability? [duplicate]

Question

Is it ever reasonable to come across an event that is very improbable and then infer that it must have been designed?

Note that my question is referring to cases where we make this inference purely from the event's improbability.

In some examples, this may seem to be the case. For example, getting three royal flushes in a row is incredibly improbable. More specifically, this equates to 1 / 2.74x10^17. Thus, design seems obviously more likely.

But we don't conclude that someone cheated purely because that sequence was improbable. We conclude this based on the facts that:

a) 3 royal flushes in a row are improbable

b) humans exist and are capable of rigging card decks

c) there have been recorded instances of humans cheating

Without b) and c), would a) be enough to conclude design? More specifically, is there a probability X such that if the event's probability was below that X, we can safely conclude design?

What about a more modest proposal? Does an event being improbable raise the probability of the design hypothesis even if it is not enough to conclude that design is what explains the event?

Answer 1

No, you cannot infer design on the basis of improbability alone. Take your card example. The odds of three royal flushes are very low, but so are the odds of any three hands with specific cards in them. You would not suspect cheating if three unremarkable hands were dealt, even though the probability of them containing their specific cards is extremely small.

There seems to be a variant of the gambler's fallacy at work here. If the probability of an event- assuming it happens at random- is p, then the probability of it happening by design is not 1-p.

The probability of design does not increase as the probability of chance goes down. What does happen is that the relative probability of design increases as the probability of chance decreases, but not the absolute probability. Therefore, to prefer the explanation of design, you need to have some independent basis for assessing that the probability of design is higher than the probability of chance.

Answer 2

Improbability of an event arising from non-design does not necessitate the conclusion of design.

See Is it ever reasonable to infer impossibility from high improbability?

Even some zero-probability events are not impossible.

Answer 3

Improbability is often used to rule out explanations but can't support alternative explanations without some other premises. If you get three royal flushes in a row, it is reasonable to rule out the mechanism of random dealing as an explanation of how that came about. This isn't the same as saying it must have been cheating because as you note, if you don't know that cheating ever occurs, that seems like something of a leap. Still the fact that you don't have an alternative explanation to random dealing does not rehabilitate that explanation. What you are left with in the absence of other information is just an inexplicable event. The astronomically improbable random deal doesn't become a better explanation just because you can't think of different explanation.

In general, if you have an event or state of affairs S, and there are two competing theories for how S came about, A and B, then a demonstration that A is astronomically improbable is a strong argument against A but not necessarily an argument for B. On the other hand, if it is a premise of the discussion that either A or B must be the case, then any evidence against A acts as evidence for B.

Answer 4

"Is it ever reasonable to come across an event that is very improbable and then infer that it must have been designed?"
No, I don't think one can. As others have pointed out, just because we can't think of another good, alternative explanation to design does not mean that this explanation does not exist. The prudent answer here would be "We don't know". At the very least, we certainly "must not" infer the design explanation from just that.

In a comment you ask "How can one rule out “non design” (I use this instead of random chance since the term chance is vague) if one can’t think of a better candidate explanation?". However, here we don't say that we rule out the design explanation; it's just that, by itself, the improbability is not enough to support that specific explanation; at least not enough to infer that it must be the correct one.

That said, this is being theoretical. If, on the other hand, there are important stakes at play, then it may be more pressing or more important to assume one explanation, instead of just saying "we don't know". For instance, this kind of use case might happen when trying to diagnose a disease, or to investigate a health issue. In this kind of case, we still "must not" infer the explanation from the improbability, but we might need to act "as if the explanation being considered is correct". However, for reference, just in case, I must say that it seems to me like the things that religions try to use this "argument of improbability" for (ex: existence of the world, or of humans) do not fall in this kind of use cases: there is no existential risk in keeping answering "we don't know" to those questions. Hence it's really a bad argument to use for those questions, as far as I can see.

Answer 5

Is it ever reasonable to come across an event that is very improbable and then infer that it must have been designed?

No. It's called an argument from incredulity or the divine fallacy.

I have given a ridiculous example of this reasoning involving a vacuum cleaner and an earthquake in another answer:

https://philosophy.stackexchange.com/a/118340/67687

From Mr. Tea:

if you were the first human exploring a foreign planet and saw a cliff face full of what looked like carvings of complex symbols and drawings, would you think back to your vacuum example and decide the patterns must be the result of random weathering and erosion?

Remember the Face on Mars? It was supposed to be evidence of intelligent design. Spoiler: It wasn't. https://en.wikipedia.org/wiki/Cydonia_(Mars)

Answer 6

I want to preface my response by quoting something from the SEP entry on probability and logic:

There is some discussion about the exact relation between inductive logic and probability logic, which is summarized in the introduction of Kyburg (1994). The dominant position (defended by Adams and Levine (1975), among others), which is also adopted here, is that probability logic entirely belongs to deductive logic, and hence should not be concerned with inductive reasoning. Still, most work on inductive logic falls within the ‘probability preservation’ approach, and is thus closely connected to the systems discussed in Section 2. For more on inductive logic, the reader can consult Jaynes (2003), Fitelson (2006), Romeijn (2011), and the entries on the problem of induction and inductive logic of this encyclopedia.

So firstly, it is insufficiently precise to say, "X is either probable or improbable." You might mean, "X has a majority chance of being true, or it does not have a majority chance of being true," which is fine, but the latter phrasing makes it clearer that "improbable" events are not in themselves surprising, since their minority-chance of coming to pass translates into them not happening especially often. What would be totally unexpected would be if pre-life spontaneously turned into life constantly, everywhere, and somehow regardless of the unstable background conditions in most such places and at most such times. That life would flicker into stability here and there, once in a while, and not so as to foreseeably endure forever (though it might, they wouldn't really know, and even though some random books in their history might claim to reveal a future state), is about as unsurprising a turn of events as can be imagined in a universe that works according to quantum field theory and general relativity in an expanding spacetime.

Hence we see that a sense of "improbability" can depend very much on framing. This is similar to how portraying an option in two different statistical ways can influence the favor people feel towards an option. "A will lead to 60% mortality," can coincide with, "A will lead to a 40% survival rate," and if one leaves it at that—leaves the audience to figure out the complementary adverse probability—one has a better shot at getting people to favor option A.

Secondly: and how exactly would such an argument proceed?

1. X had a low probability of happening.
2. X happened.
3. Therefore, X was (probably) intended to happen (by an agent/agents with sufficient physical power to actualize this intent).

From a deductivist vantage regarding probability logic, one is minded to amend (1) to (4):

4. If X was not intended to happen, X had a low probability of happening.

Yet by itself this is far from good enough for present purposes, because we know from many and varied experiences that it is often not enough that we intend something to happen, such that it does happen because we intended it to. Plans go awry, our walks down our paths lead us astray, plenty often. When the action in question is (relatively) simple, like blinking or preparing a small meal, our chances of "perfect" success are much greater (indeed, it is not hard to conduct a "perfect" blink). But it is not the strength of our intent that decides these matters alone: the ease of success is a function more of the type of those actions than of the degree of willing involved. Again, it would be pretty difficult to vary one's degree of willing with respect to blinking; in this context, "Do or do not: there is no try." But that is consistent with other contexts admitting of trying.

Now, we should reformulate (4) as (5):

5. If a powerful enough being did not intend that X, then X had a low probability of happening.

Of course, the probability that a powerful enough being could not only intend X but certainly bring X about, is actually 100%, trivially. That is, the probability of X happening if a powerful enough being so intended is 100%. But obviously this does not negate the possibility of X happening unintentionally. Anything could be "explained" as, "It was willed by a sufficiently powerful being," if that were the case, which it is not (and abstract occasionalism aside). But this would not be what we were looking for when we asked about explanations, which we frame in terms of mechanisms and processes involving mathematically precise ranges of functions. We do this even when, like Leibniz, we view intercausality as a distorted or fictitious conception of the natural order, because Leibniz would not hesitate at all to delve into the infinitesimal calculus, with all its risk of chaos and paradox, for the sake of a greater understanding of things.

Answer 7

Note that my question is referring to cases where we make this inference purely from the event's improbability.

There is never a case where you make this inference purely "from the event's improbability" because if you infer design then you already have some notion of design(er).

In your case

• if "getting three royal flushes in a row is incredibly improbable" and you do not have a notion of cheating in your mind (you are for example young trusting child) then you cannot infer cheating.
• but the moment you are aware of concept of cheating you can infer cheating but it is not purely from "getting three royal flushes in a row" improbability.

Answer 8

Derek Parfit in his beautifully written article Why Anything? Why This? distinguishes two scenario's:

Suppose first that, of a thousand people facing death, only one can be rescued. If there is a lottery to pick this one survivor, and I win, I would be very lucky. But there might be nothing here that needed to be explained. Someone had to win, and why not me? Consider next another lottery. Unless my gaoler picks the longest of a thousand straws, I shall be shot. If my gaoler picks that longest straw, there would be something to be explained. It would not be enough to say, ‘This result was as likely as any other.’ In the first lottery, nothing special happened: whatever the result, someone’s life would be saved. In this second lottery, the result was special, since, of the thousand possible results, only one would save a life. Why was this special result also what happened?

The Big Bang, Parfit argues, was like this second lottery. For life to be possible, the initial conditions had to be selected with great accuracy. This appearance of fine-tuning also needs to be explained.

For life to be possible on Earth it needs to circle around the Sun in a narrowly defined habital zone (sometimes called the Goldilocks zone). The chances of that being the case are slim. However, with the growing understanding of the vastness of the Universe and the detection of exo-planets, the odds were no longer that astronomical. In other words, it became more like the first lottery and no explanation is needed.

The Big Bang with its appearance of fine-tuning can be explained in two ways: God or the assumption of many universes with different initial conditions.

Answer 9

I think the biggest issue with this kind of comparsions is that it assumes perfect knowledge about the possibilities and the process as to how one of those possibilities is chosen.

If you watch a game of poker and one player gets three royal flushs in succession you could reasonably infer that he is cheating. Yes there is a very slim chance of him being actually that lucky. In a regulated poker match you would still have to proof his cheating though to convict him, but it is reasonable to start the investigation with the assumption "that" he cheated and instead focus on finding out "how" he cheated. (In an illegal backroom poker game he would get "reasonably" beaten to a pulp without any investigation though).

But for poker you know the rules, you know that each card in the stack is unique, you know that the cards are shuffled by a perfectly random shuffling machine and you know that no card drawn is influenced by the other cards being drawn (except that they can't be the same).

But now you watch me and my friends play a card game. You don't know the rules, you don't know the deck. You only see the cards that are drawn. You can do estimations about the probability but your lack of information is making that more a guesswork than anything else.

Maybe the deck only has one color. Maybe the deck is shuffled in a very special way that makes previous hands more likely to appear again. Maybe the king and queen are quantum linked and are always drawn in succession. You don't know.

Yes we can reasonably infer design from improbability, but we can't if that calculated improbability is mostly based on guesswork.

Answer 10

You are overlooking an important thing about improbability: The longer time, the more times you try, the more probable is your improbability!
You talk about a deck of cards, and a fairly short number of games giving three royal flush.
For simplicity, let's throw ten dice. The probability of all ten dice showing 6 is 1:6^10 which is 1/60466176 - one out of 60 millions.
But now throw these ten dice a billion times. The probability of not having all of them show 6 at least one time is 1:((1 - (1/60466176))^1000,000,000) = 1/15220774 - one out of 15 millions.
Improbability is very depending on how many times something is tried.

Answer 11

Suppose we come across a Dyson sphere in outer space. Humanity has no experience designing such structures, nor the advanced technology and physics knowledge likely required to construct one. We have no documented cases of anyone designing a Dyson sphere, nor would any human possess the expertise to claim, “Yes, it was designed, and here’s how it’s done step-by-step.” So, how could we explain the existence of a Dyson sphere if we encountered one? Here are three possible hypotheses:

• H1: It was designed by some unknown intelligence (certainly not us).
• H2: It formed by chance, through phenomena such as quantum tunneling.
• H3: It emerged spontaneously from nothing, without any cause.

For H2, we might argue that its probability, based on any reasonable estimates from experience, is exceedingly low. However, in principle, it remains possible. In a sufficiently large (or infinite) universe, multiverse, or megaverse, such events might happen somewhere — unless future advances in physics definitively rule it out. Given our current understanding, though, we cannot exclude H2. You could think of this alternative as the Anthropic Principle applied to Dyson spheres.

Now, if P(H2) is extremely low, would it be reasonable to conclude that P(H1) is high? This depends on whether we consider a sufficiently large megaverse or multiverse in which even Dyson spheres might arise by chance. If so, our local probability estimates would miss the mark, as they consider only our universe’s statistics, not those of the entire megaverse. If, however, our local statistics are accurate, then P(H2) does appear extremely low, suggesting either H1 or H3 is likely. Since H3 also seems statistically improbable from experience (how often do we observe Dyson spheres appearing around stars or any other megastructure for that matter out of nothing?), H1 could seem the most reasonable.

In conclusion, it ultimately depends on whether one chooses to rely on local statistics or to assume the existence of vast multiverses, which could provide ample opportunities for unlikely events—such as the formation of colossal megastructures by chance or spontaneous occurrences without cause. If you're not inclined to either accept or reject assumptions like these, then it seems to me that the most reasonable stance would be to remain agnostic about the probabilities.

Answer 12

Does an event being improbable raise the probability of the design hypothesis even if it is not enough to conclude that design is what explains the event?

No event is probable or improbable in itself. Instead, for each event there are many (often infinitely many) models of how it was caused. and for each model, if the probability is shown to be low, it's plausibility is lowered. This does not raise any other plausibility of any other model.

The scam that is ID is based on the Strawman fallacy that "science" only has one model for a given event, and that this "single model" has a realistically calculable probability, and that the only alternative to that "single model" is design.

The truth is that while science often and necessarily proposes models for past events, those are not exclusive single models, but just picked suggestions out of a nearly infinite field of other possible naturalistic models. Also typically there is no useful way of calculating the probability of such models, as this would require precise observations of earth at the time, which we do not have.

So even if a suggested naturalistic model for any event was wrong, it does not automatically mean a design model must be true, there are near-infinite naturalistic alternative models to consider, always.

And even if a given scientists suggests a model, it's probability cannot be calculated as suggested by ID pseudoscience.

All science can do (and does) is to attempt to reproduce environments and see if similar things happen today. And if this does not work, try our any of the infinite variations, learning along the way. There has been no reason to give up on this process anytime in the last 3000 years. Nor will there likely be any reason to give up on this process in the next 3000 years.

Infinity is a long time, and humanity still has roughly that much time to try out and exhaust naturalistic models for all the events in the universe that still lack a proven model. And with all the knowledge gained in the process, we can build bridges and cards and smartphones and satellites. But ID as a political movement would rather invest energy in teaching Christian lies in schools.

Intuitively, consider the Blackjack card game where players try to reach a score close to 21 that is higher than the score of the dealer. Imagine this with 1000 players and one dealer. So we look at player one. And let's say player one stands at a score of 14. That's a low probability of winning. Lower than say at 20. But standing at the low probability of 14 does not mean that the dealer has a blackjack (or would win). It also does not mean that the dealer would equally beat all the 999 other players. It just means the probability of the first player winning is low.

ID tries to convince people that the dealer must win because the chance of the first player to win is low, and there are no other players. And ID uses strawman arguments as the initial player.

This is why ID is considered pseudoscience not even worth investigating. (That it is only pushed by an organization that is a sockpuppet for by Christian fundamentalists and Creationists does not help, but certainly that is not the only reason to reject ID)

Answer 13

Is it ever reasonable to come across an event that is very improbable and then infer that it must have been designed?

In general, "no". In order for a proposed cause to be considered viable, it should also possess sufficiency.

That said, "sufficiency" tends to cut against non-design causations. For example, random processes have never been shown to produce meaningful information (e.g. the works of Shakespeare, or a functional biological organ) from "scratch". Conversely, intelligent agents manipulate the world in a way to actualize extremely improbable outcomes all the time, so while the answer here is technically "no", it's unclear what situation could arise in which an intelligent agent (of suitable capabilities) would not be a sufficient cause for some improbable event.

For example, getting three royal flushes in a row is incredibly improbable.

Here we see the need to introduce another factor: specificity. Getting any specific three hands of cards is extremely improbably. Nevertheless, if I sit for three hands at a poker table, an extremely probably event will be actualized. "Entropy" plays a role, but what we're really trying to measure is "randomness". This is a whole, genuine field of study that can be qualified, but to keep things simple, "interesting" outcomes are both specified and complex.

A deck of cards adds some complications, so consider, instead, a series of coin flips. The outcome of all-heads is highly specified, but not complex; the most likely explanation is neither chance nor design, but necessity (namely, your coin probably isn't fair). A series of perfectly alternating heads and tails is more interesting, but still not complex enough to strongly suspect design rather than necessity. (For a more "real world" example, crystal growth is an example of necessity.) If, on the other hand, there is no discernible pattern, then the outcome is highly complex but is not specified, and is likely a result of chance.

Now, if you flipped a coin, and the flips, when "read" as binary representation of ASCII, spelled out the Declaration of Independence, you're dealing with a result that is both complex and highly specified; design is now a much more likely explanation than chance. Similarly, any three "typical" hands of poker are extremely improbable, but they are not specified. Three royal flushes are both unlikely and specified, therefore making design (cheating) a more likely cause.

Now... here's where things get interesting, because this is the point at which the infinite monkey theorem is usually invoked (including by several existing Answers here!). While technically correct, the theorem only works in the world of infinities, and only in certain conditions. To understand, however, it is necessary to clarify the nature of our unusual event.

Going back to the poker example, if Bob sits down and is immediately dealt three royal flushes, that's cause to suspect chicanery. If, on the other hand, Bob has been playing poker for 10¹⁶ hands, then it is much less "surprising" if, at some point, he was dealt three royal flushes in a row.

I'll circle back to this in a moment, but it's leading us to another point:

Is there a probability X such that if the event's probability was below that X, we can safely conclude design?

Yes. Often called the "universal probability bound", if we estimate the total number of particle interactions that have occurred in the history of the universe (usually cited as around 10¹⁵⁰), then anything unlikely to occur given that many "chances" is unlikely to have occurred by chance. (Obviously, the number of poker hands that can be played is much, much lower than this number. More "realistic" estimates use a bound of 10^-50 as the threshold for excluding chance as a plausible explanation.)

The oft-cited solution is to posit that existence is, in fact, infinite (a.k.a. the anthropic principle). The problem with this, besides being a wholly philosophical supposition with no concrete evidence to support it, is that it doesn't actually work in practice... and this is where we come back to the monkeys and how theorems about infinity don't work in a finite universe.

Let's use another Answer's Dyson sphere as an example. It's theoretically possible (albeit far, far below the universal probability bound) for such a structure to spontaneously materialize... just as infinite monkeys will eventually produce the complete works of Shakespeare. However, in an infinite universe, there will be infinitely many humans that have not encountered a Dyson sphere. The probability that, by chance, our universe happens to have both humans and Dyson spheres is itself unreasonable. In other words, the anthropic principle cannot defend chance as an explanation. The only plausible explanation for our universe happening to have both humans and Dyson spheres is either design or necessity.

Now, I can imagine no argument that makes the existence of Dyson spheres necessary in order for there to also exist humans to wonder about them. Moreover, it's difficult to imagine even the many unlikely forms of life that we do observe being necessary for observers to wonder about them.

To be fair, technically the humans in the universe with spontaneously generated Dyson spheres could still fall afoul of the anthropic principle, but this requires belief in infinite universes, for which we have no scientific evidence. Such belief is purely philosophical. Worse, because the probability of Advanced Aliens existing in an infinite multiverse is non-zero, it's actually more likely that a universe contains Dyson spheres because someone built them. In fact, if the probability of Advanced Aliens being able to create entire universes is non-zero, it immediately becomes more probably that any "interesting" universe is, in fact, designed.

Answer 14

I would like to add another perpective to the quite varied and interesting set of answers already present, which, i believe, might help a bit by presenting another angle to this topic.

I do not believe that one can infer design from improbability alone, here is why:

Imagine that Alice and Bob are two reporters, each tasked with writing an article about how it feels to win the Grand Lottery. The chances of winning are 1 in 10^20.

• Alice buys a ticket, picks her numbers, wins and then proceeds to write her article.

Now, this seems exceedingly unlikely given the probability and i would be very skeptical of the truth of her article, if i ever were to read it.

• Bob, on the other hand, finds a winner or waits until someone wins the Grand Lottery, visits and interviews them and proceeds to write his article.

The underlying probability of winning here is just as in Alice's case, but it appears vastly more likely to have happened. Generally, i would say this article seems believable, especially since it will only ever be written if someone, anyone wins.

Now, a problem i see arise most often with questions such as the above is, that people try to reason about a situation similar to Bob's, looking back at something that has already happened, but apply probabilities the way they are used in Alice's case, looking forward to something that is going to happen, but with a very specific outcome in mind.

Most of the time the "improbability argument" pops up, we are clearly in Bob's situation. Something has happened, then somebody tries to explain its specialness and somehow ends up acting like we are in Alice's case with respect to the probabilities. But the question for an explanation can only be asked if there was a winner. And the numbers are only special because they came up. If other numbers produced another winner, the same question would be asked about those.

Given the contexts in which the "improbability argument" is usually used, we are woefully unequipped to assign an informed propability to the equivalent of "Bob's situation", even if we were to know the minutae of "Alice's" probabilities. We know even less about possible alternative explanations.

Addendum:

I tried finding something more mathematically rigorous on this topic, but so far i only found this interesting blogpost, which might help illustrate the point a bit further via a rather simple example using Bayes Rule. The two most relevant statements from it are:

Backward Probabilities require information from all possible causes whereas Forward Probabilities do not

and

Backward Probabilities can be drastically different from analogous Forward Probabilities

If we accept the reasoning in this example, then the given - improbable but not impossible - forward probabilities are never sufficient to conclude something about the backward probabilities in and of themselves. This includes the validitiy of any other possible explanation besides chance, like a designer - unless we were vastly better informed about all the possible explanations' respective base rates, i guess. ;)

Answer 15

Frankly, yes. And I'm speaking as an atheist. My first major in college was physical anthropology.

If something created us, it wasn't God, it was sufficiently advanced technology that's indistinguishable from magic.

Anyone before us would definitely have evolved millions of years beyond us. Creating a universe as a garden is exactly the kind of thing you'd expect them to do.

When they appeal to probability and long times, they remind me of someone afraid to talk about the elephant and the parlor when it's slapping them in the face with its tail.

They exquisitely go into the details of how this astounding complexity can occur naturally, but they never address WHY it occurred.

It's like if, over thousands of years, water erosion turned a granite rock into the statue of David.

Yes, it CAN happen but why did it?

Space aliens!

Answer 16

My two cents follows.

Suppose we are shown a recording of what we are told is a very unusual event in a game, similar to all players getting a royal flush in poker. We might think the cards were loaded, that we've misunderstood the rules of the game, that the game's been going on for some time, that the video is faked, etc.. Do we have a way to choose between those scenarios, if all we have is the recording?

Besides which, I like all the counter arguments here, and probably think that there is nothing remarkable about being able to make observations: that the universe supports them is one observation among very many, more general but no more special than any other.

The question, as I understand it, not your question, needs more clarity.

Answer 17

The question is about reasonableness:

Is it ever reasonable to come across an event that is very improbable and then infer that it must have been designed?

That is, the question does not seem to be asking "If something seems very improbable, can you ever say with absolute confidence that the improbable thing didn't happen?" - the answer is no, given only that information, you can never have a higher confidence than that probability :)

So I take "reasonable to infer" as "a reasonable person, given the balance of probabilities, would not find that any alternative hypotheses need be considered".

As for "design", I take that to mean "created with a deliberate goal in mind".

The canonical example of this is the pocketwatch.

In the absence of a designer, who had the deliberate goal to create a timepiece, the existence of a pocketwatch is vanishingly improbable.

So it is entirely reasonable in this case to infer a designer, given no more evidence than the existence of the pocketwatch. There is no plausible alternative hypothesis: we could appeal to magic and say someone just waved a wand and said "let there be timepieces" and there was one, but that's not plausible. The balance of probabilities does not support it, even if someone writes in a book that it's how it happened.

Now say we have a pocketwatch, with an English textual label engraved on it, saying "Designed By Fred". This watch is demonstrably similar to all other pocketwatches by the same designer. It has a certificate of authenticity, a well-documented chain of custody, and if you ask Fred himself, he'll tell you he remembers making that specific watch, and will be able to point to some remembered toolmarks, and correctly telling you where you can find others in the watch's internals. You can observe Fred designing and creating other watches.

It would be an exceptionally unreasonable position to claim that this wristwatch was never designed, and that the book about the wand is correct instead.

---

The distinction between a pocketwatch and a mouse or a moon, is that in the latter cases, it is exceptionally unreasonable to claim that any living thing or planetary system was intentionally designed. Not only is there a plausible alternative hypothesis for these things (evolution, gravitational accretion) but it's the overwhelmingly more likely hypothesis.

All that supporting evidence that we saw for the pocketwatch's design being Fred's, is similarly present for these, and vastly more, and it all points to the "design" being created through evolutionary/gravitational processes... which require no "deliberate goal".

Now, we could argue "there might be an intent guiding the hand of evolution/universe formation, so gently that it cannot be statistically detected above the level of random noise in any form of analysis, but having an immense effect over millions of years." That is, that evolution might have a subtle artificial-selection element to it.

That hypothesis cannot be disproven, but neither can the null hypothesis of no guidance, or the extreme hypothesis that there are two, three, or even a million different intents all subtly battling to direct life's development in different ways.

So just like if there was a book about someone waving a wand and saying "let there be life", so the idea of "let life be directed from random chance, but in like really very small amounts, by precisely N entities", is not a plausible hypothesis. A reasonable person would not accept that hypothesis as having significant weight in the balance of probabilities.