The Problem Making Clear the Problems

[James William Hall]

I submit the following in the “General Discussion” category to allay any pretentions on my part.

“How” and “why” are both interrogative words, but they answer different questions. “How” answers questions like “By what method?” “To what degree?” “In what condition?” and many more. Whereas, “Why” answers questions like “For what purpose or reason?”

Does science ever answer “Why?” to physical problems like the universe expanding and apples falling? My impression is that otherwise very smart people confuse themselves and their admirers by using first one, then the other word, as they go along doing their physics. Now every time I read a post in Physics Forums I look for the two words as in, “Why is the universe expanding?” Come to think of it, if you were guaranteed an exact answer to "How" or "Why" apples falling, universe expanding, or others in the Forum, which would have precedence in your opinion?

 

[fresh_42]

I think Feynman's answer to this question cannot be given any better:

 

[James William Hall]

If you read the definitions of "How" and "Why" above, it would seem that Feynman is mixing them up too which was my point, no matter that Feynman is a hero of mine (even if snippy with lesser mortals). Perhaps along the timeline we are merging the "How" and "Why" into one interrogative out of convenience.

 

[fresh_42]

Regardless of the Feynman example, or even of your two selected interrogations. Every question has always the problem, that it does not say what an allowed, acceptable, satisfactory answer will be. Neither how nor why nor any other word includes this crucial information. You can easily observe this phenomenon on PF. It is the origin of at least half of the posts - this thread included! It is also the reason why we have the "A-I-B" categories. But this is more of a helpless attempt to sort the questions than it is a real advantage.

 

[russ_watters]

If you read the definitions of "How" and "Why" above, it would seem that Feynman is mixing them up too which was my point, no matter that Feynman is a hero of mine (even if snippy with lesser mortals). Perhaps along the timeline we are merging the "How" and "Why" into one interrogative out of convenience.

Maybe the problem is with the English language? Or with the applicability of the question? Nobody would ever confuse "where" and "who".

 

[russ_watters]

Ok, I got a "skeptical" and these questions annoy me, so I'll just run through the whole logic rather than try to walk everyone through it.

The simple answer to the OP is No, science does not answer "why" questions.

What annoys me about this question (and it comes up frequently) is it is sometimes a misunderstanding of or failing of English or a misunderstanding of science, but more frequently it is an attempted "gotcha" of a limitation in science.

The Feynman question was "what", and then "why", but Feynman only addresses half the issue. Feynman was addressing the "rabbit hole" of questions, where there can be a lot of potential answers with varying potential depth and it isn't always easy to know what will satisfy the questioner.

By the way, this is usually done with "how", as the OP suggests, so I'll go with that:

Consider:
Q1: How did you go to the grocery store?

Possible answers:
A1: Main street.
A2: In my car.

"How?" asks "by what method?". Now obviously either of these answers can be followed-up and dissected with significant depth. How does a car work? How did you get your car into the grocery store?

Now consider:
Q2: Why did you go to the grocery store?

Obviously:
A3: To get food.

But that question also has follow-ups and depth. Why did you want food? I'm hungry. Why are you hungry? Because my stomach said to my brain that it needs food. Why did your stomach say to your brain that you need food? Etc.

Now, you might notice that the rabbit hole of "why" questions actually morphs into improperly labeled "what" or "how" questions. "Why?" asks: "for what reason?" and the word is intended to investigate an intelligent decision. When applied to people, "how" and "why" questions have obvious distinctions because we know that "why" is asking us as people what our reason is for deciding to do something and "what" is asking the method we used to execute our decision. That's why how and why you went to the grocery store are two very different questions whereas how and why is the sky blue are the same question, perhaps improperly combined.

So, no, science doesn't answer "why" questions because it isn't about making decisions to do something -- unless of course the answer is "god wanted it that way".

 

[James William Hall]

Coincidently, I am up to page 10 reading, “The Grand Design”, by Stephen Hawking and found this:

“To understand the universe at the deepest level, we need to know not only how the universe behaves, but why.

Why is there something rather than nothing

Why do we exist

Why this particular set of laws and not some other

This is the Ultimate Question of Life, the Universe, and Everything. We shall attempt to answer it in this book. Unlike the answer given in The Hitchhiker’s Guide to the Galaxy”, our won’t be simply “42”.

 

[Evo]

There is a reason we don't ask a lot of "why" questions in science...it becomes too speculative and too philosophical. Reminds me of the cartoon where the Dog is asking "Why am I here?" "What is my purpose?" Then it said "I'm wearing my Hartz Reflective Collar". I just about died laughing. I clipped that one out.

This thread is just about over.

 

[BillTre]

The simple answer to the OP is No, science does not answer "why" questions.

I would agree with this with respect to what might be called the physical sciences (like physics and chemistry, as usually understood).
There situations, studied by science, which have answerable "why" questions.

These are situations, overlooked by many physical scientists, that involve the behavior and properties of independent reproductive agents. This would be things like living entities and particular (lab based) chemical systems with replicative properties.
Competing replicating systems will evolve capabilities and goals that enable them to better survive and proliferate. This is their ultimate answer to why, but there are many at intermediate levels.
Why is there a heart?
To pump blood, to distribute and remove chemicals, to survive and reproduce.
This is a lot like "why is something engineered in a particular way?" In order to achieve the goals of the job. The ultimate job of replicating entities that have undergone competition, is to reproduce.

Here is a longer post in a different thread I wrote on this.

So, I think a no answer to this question should be qualified by limiting the answer to the study on non-animated things.

 

[russ_watters]

...agents...

So, I think a no answer to this question should be qualified by limiting the answer to the study on non-animated things.

I'm good with that. One issue here is that it isn't easy to identify agency, which to me is the key difference between why and what or how. So it isn't easy to draw the line between what is "animated" and what isn't.

A deterministic physicist or chemist might even believe there is no such thing.

And personally, I think QM just makes it worse.

 

[BillTre]

One issue here is that it isn't easy to identify agency, which to me is the key difference between why and what or how. So it isn't easy to draw the line between what is "animated" and what isn't.

Thanks for bringing this up.

It seems to me kind of obvious that living things are agents since they are autonomous and take care of their own requirements for continued functioning. However, living things are (now-a-days) well studied and understood to a high degree of detail. Two hundred and fifty years ago, many (small, slow moving, and/or slow growing) things would not have been seen as alive.
Similarly, a lack of detailed knowledge could limit identifying agents.

Your comment inspired me to google agents, which is a nightmare (too many other "agents").
But I found a article from 1996 (here) that tries to define agents in a way not just limited to software.

Here is the definition they settled on:

An autonomous agent is a system situated within and a part of an environment that senses that environment and acts on it, over time, in pursuit of its own agenda and so as to effect what it senses in the future.

One way of clarifying the boundaries of this definition is by looking at extreme cases. Humans and some animals are at the high end of being an agent, with multiple, conflicting drives, multiples senses, multiple possible actions, and complex sophisticated control structures (minds [Franklin 1995]) . At the low end, with one or two senses, a single action, and an absurdly simple control structure (mind?) we find a thermostat. A thermostat? Yes, a thermostat satisfies all the requirements of the definition, as does a bacterium. Strange things sometimes happen at the extremes. Espousing a definition entails these risks.

and they came up with an agent taxonomy which recognizes biological entities ( ):

I was going to say that the viruses should be in the biology group, but I guess they meant software viruses (duh).

Not sure that makes it any easier to ID agents out in the world, but its nice to have a definition to get started with.

 

[russ_watters]

Thanks for bringing this up.

It seems to me kind of obvious that living things are agents since they are autonomous and take care of their own requirements for continued functioning. However, living things are (now-a-days) well studied and understood to a high degree of detail. Two hundred and fifty years ago, many (small, slow moving, and/or slow growing) things would not have been seen as alive.
Similarly, a lack of detailed knowledge could limit identifying agents.

Here is the definition they settled on:
"thermostats"

I have indeed used a thermostat in the past to illustrate what's at the absurd bottom of the rabbit hole that is trying to define life (or intelligence/agency). I submit that 250 years ago people also believed that there was a "special sauce" that made life - such as human life - somehow different from non-life. I don't think that's the prevailing view anymore. If AI is just a complicated computer program and life a complicated chemical reaction, then where to draw the line on agency is arbitrary and meaningless.

So my agreement is only half: I agree with what is on either side of the line. I don't agree with where the line is drawn. And I recognize the implications of where I would draw it make people uncomfortable, even if (in my estimation) the logical basis is better.

 

[jbriggs444]

Not sure that makes it any easier to ID agents out in the world, but its nice to have a definition to get started with.

My personal understanding of an "agent" is when we model a sub-system which does exciting and non-linear things by putting a label on a black box. It is too complex to model completely, so we hide the complexity under a layer of abstraction and resign ourselves to merely characterizing the behavior of the box.

 

[Jarvis323]

I'm not a fan of Feynman's answer. Of course you can always keep asking deeper and deeper questions. Most people know and understand that from gradeschool. When he finally finishes sidetracking, he addresses the question by just saying if you were a physicist I could explain some things but you're not so you wouldn't understand. It gives the impression that he was trying to dodge the question. And in the end he did. Yet, he could have said a lot to help us understand what has been learned about magnetisim, the mysteries that remain, and why it's so difficult to further our understanding.

 

[Jarvis323]

Regardless of the Feynman example, or even of your two selected interrogations. Every question has always the problem, that it does not say what an allowed, acceptable, satisfactory answer will be. Neither how nor why nor any other word includes this crucial information. You can easily observe this phenomenon on PF.

I'm not a fan of the absolutist mentality which demands the questions have decisive and authoritative answers. People seek understanding by asking probing questions.

 

[fresh_42]

I'm not a fan of the absolutist mentality which demands the questions have decisive and authoritative answers.

Questions that aren't meant to be answered are called rhetorical, or simply gibberish. It has nothing to do with authority, it has to do with the problem to match an expectation. This is neither authoritative nor decisive. Without knowing the expectation behind the question, any answer will be meaningless.

 

[Jarvis323]

Without knowing the expectation behind the question, any answer will be meaningless.

I disagree. People asking questions don't know what they don't know. You can safely assume they want to learn, and approach the question as an opportunity to communicate to them what you know about something.

 

[fresh_42]

People asking questions don't know what they don't know.

This is patronizing!

 

[russ_watters]

I disagree. People asking questions don't know what they don't know. You can safely assume they want to learn, and approach the question as an opportunity to communicate to them what you know about something.

I agree with you and think this is a significant problem on PF; people spend too much time/effort questioning or criticizing the OP when a simple "starter answer" would be more helpful and could move the conversation forward instead of getting it stuck.

 

[Ygggdrasil]

With regard to the discussion on how & why questions in biology, biologists often make a distinction between proximate causes and ultimate causes (for more information, see the famous essay "Cause and Effect in Biology" by Ernst Mayr or this updated review on the topic from 2011). Here's a good excerpt from the introduction from the 2011 review:

In November 1961, Ernst Mayr, Harvard evolutionary biologist and architect of the modern evolutionary synthesis, published what was to become a classic paper on causation. Mayr’s “Cause and effect in biology” (1) had a massive influence by shaping how most contemporary biologists understand causality and was a major contribution to the philosophy of science (2). In this article, Mayr distinguished proximate from ultimate causes. A proximate cause is an immediate, mechanical influence on a trait: say, the influence of day length on the concentration of a hormone in a bird’s brain. Ultimate causes are historical explanations; these explain why an organism has one trait rather than another, often in terms of natural selection. Although the proximate-ultimate distinction had been made earlier [e.g., (3)], it was Mayr’s article that led to its widespread acceptance.

Mayr argued that “functional biologists,” such as physiologists, trade in proximate causes because they are interested in showing how systems work. In contrast, evolutionary biologists trade in ultimate causes, because they are interested in why history has produced one system rather than another. The example that Mayr used to illustrate this distinction, avian migration, drew on his early career as a naturalist. Mayr emphasized that, to fully comprehend migration, we need to understand both why birds migrate (its selective advantage) and how they migrate (how they time migration, how they navigate, etc.). How-answers complement why-answers, and vice versa.

https://science.sciencemag.org/content/334/6062/1512

Of course, this framework is very specific to biology; while evolution is a fundamental principle underlying most of biology, it is not applicable to understanding phenomena in other branches of science like chemistry or physics. Instead, maybe one could broaden the idea of proximate vs ultimate causes a different way: Proximate causes deal with the specific mechanisms behind an observation or phenomenon, where as ultimate causes try to generalize those mechanisms to fit into a broader framework of understanding a specific topic (e.g. natural selection and evolution in biology).

For example, let's consider another biological example on most of our minds, the evolution of new Coronavirus variants. The proximate causes (how did new variants evolve) would be mutations to the gene encoding the spike protein that increase the affinity of the spike for the ACE2 receptor or decrease the binding of the spike to neutralizing antibodies. Of course, these proximate causes are very specific to SARS-CoV-2 (for example, even a similar phenomenon like the evolution of new influenza strains comes about in very different manner [recombination of neuraminidase and hemagglutinin proteins from different virus strains in addition to mutations]). However, the ultimate causes behind the phenomena (why did the variants evolve) is the same in both cases (greater evolutionary fitness) and applicable to understanding a wide variety of phenomena in biology.

Let's consider an example in chemistry, say a reaction between two substances. Discussion of proximate causes (how does the reaction occur) includes discussions of reaction mechanisms and the interactions of molecular orbitals and valence electrons that would be specific to the compounds in question. Discussions of ultimate causes (why does the reaction occur) would center around the themodynamics of the reaction, showing that the reaction occurs because the products have a lower free energy than the reactants (a principle that is broadly generalizable to most chemical reactions). The idea of minimizing energy is even generalizable to be an ultimate cause of many phenomena in physics (e.g. why does a hanging rope adopt a specific shape or why does an apple fall).

Perhaps this disctinction (specific causes vs generalizable causes) might be a productive means to extend how some biologists' look at how & why questions to questions in other areas of science.

 

[Stephen Tashi]

I agree with you and think this is a significant problem on PF; people spend too much time/effort questioning or criticizing the OP when a simple "starter answer" would be more helpful and could move the conversation forward instead of getting it stuck.

There is a difference between criticizing the question and criticizing the questioner. Remarks that imply the original poster is lazy, ignorant, etc. aren't helpful. Remarks that criticize the original post as being ambiguous or vague can be useful if they are specific. It isn't helpful to say "Your question is ambiguous" or "You need to clarify you question" without at least giving an example of different interpretations of the question or particular points that are unclear.

(Also, we need not assume that the purpose of an original post is to ask a question.)

I agree that starter answers are useful in the case where a plausible specific meaning of an original post can be guessed even though the original post isn't worded clearly.

In topics of probability and statistics, it's typical for an unclear original post to get several starter answers and then we get a lively debate about which are correct - and never define a mathematical question precisely.