Hmm. Something like this?

1. We've carved up the world conceptually. Word-concepts apply to a variety of things.
2. We encounter something new through exploration.
3. This something has the manifest image of a pre-existing word-concept and so is something we could at first sight seemingly cover with the existing word-concept.
4. However, we discover that this thing has a scientific-image that is different from that which underlies the previously known iterations of the manifest image.
5. Therefore we have two options:
a. Extend the pre-existing word-concept to cover all versions of the manifest image and restrict the use of differential vocabulary to distinguishing between the two scientific images which underlie the manifest image.
b. Add a new word-concept in order to distinguish the manifest image by variations in the underlying scientific images.

Neither. It is ambiguous. "Snow" is being used as a phenomenological category rather than as the name of a substance.

It is like stepping on a very earthbound beach and saying that "the sand here is so white", without knowing that it is actually composed of the pulverized remains of mollusk shells (calcium-based rather than silica-based). On Earth, we use the name "sand" for any granular rock-based material (of a given size), without regard to its substance (silica, calcium, etc.); and we use the name "snow" for a given type of water-based crystals, merely because we are not used to seeing snow of a different kind (unlike the case of sand). But if we had snow of different origins here (and we probably have, but I have no idea of their kinds), the word would be "naturally ambiguous" (like sand) rather than just, er, "philosophically ambiguous" :D.

Therefore we have two options:
a. Extend the pre-existing word-concept to cover all versions of the manifest image and restrict the use of differential vocabulary to distinguishing between the two scientific images which underlie the manifest image.
b. Use different word-concepts to distinguish between variations in the scientific images which underlie the two more subtle distinctions in the manifest image. — John Doe

There is the manifest difference in temperature ranges often found on Earth where C02 ice sublimates into a gas, while H20 turns into a liquid.

Why do you use the word "image" for the scientific understanding? I get the manifest image, because it captures the notion of appearance for humans, given how visually dominant most of us are. But science is more abstract. Is "image" another word for description or model?

But if we had snow of different origins here (and we probably have, but I have no idea of their kinds), the word would be "naturally ambiguous" (like sand) rather than just, er, "philosophically ambiguous" :D. — Mariner

I was going to relate that Eskimos have 20 different words for snow, but now I see that's a controversial claim, involving a dispute over linguistic relativity.

But yeah I imagine that if we grew up on Mars and encountered both water and carbon dioxide snow, we might have two related but different words for snow.

Well there's a big literature on this stuff that I don't know terribly well, but in my idiosyncratic use I'd want to say 'scientific image' is how something looks from the scientific point of view. And I would think that, just as this chair looks different depending on the light, the position of my body, etc., the scientific image of something can change depending on how you're capturing it in various descriptions, models, applications, theoretical explanations, etc.

the scientific image of something can change depending on how you're capturing it in various descriptions, models, applications, theoretical explanations, etc. — John Doe

So you don't accept Nagel's view from nowhere as the scientific image?

False. It's a snow-like substance, but not snow.

snow — Marchesk

Nice fluffy 6 sided crystals of H2O = snow. What do the crystals on the south pole of Mars look like? Cubes.

Nice fluffy 6 sided crystals of H2O = snow. What do the crystals on the south pole of Mars look like? Cubes. — Bitter Crank

Didn't know that. So would it be false to say that every snowflake on Mars is unique?

H2O snow vs CO2 snow? We can have different ices and salts, why not snows?

Apparently so:

"What do you know, snow is white on Mars!"

But the snow is made up of carbon dioxide crystals, which melts into CO2 liquid when the temperature is low enough, which it can be near the Martian poles. Is your statement true or false? — Marchesk

As usual it depends on what the statement means. Conventionally, "snow" refers to H2O, so the astronauts are not seeing snow, they're seeing something else (dry ice, as it happens). Nonetheless snow (H2O) is white on Mars as it is on Earth. So the astronaut's statement is true.

But note in your second sentence above that you use "snow" in a different sense to refer to what the astronauts observed (dry ice). On that meaning, the astronaut's statement is true as well, since dry ice is white on Mars.

The main issue here is to avoid equivocation.

False. It's a snow-like substance, but not snow. — Sapientia

As usual it depends on what the statement means. — Andrew M

The snow is white is true iff the snow is white.

But then that depends on what we mean by snow, since snow-like stuff can have different chemical compositions.

But if we had snow of different origins here (and we probably have, but I have no idea of their kinds), the word would be "naturally ambiguous" (like sand) rather than just, er, "philosophically ambiguous" — Mariner

Ordinary language use is ambiguous and thrives on that fact. Formal language is the attempt to remove ambiguity so as to provide the kind of certainty and absoluteness demanded by logic.

Ambiguity can never actually be removed. It can only be constrained to an arbitrary degree. Definitions always remain “open to interpretation”.

But formalism does apply generic syntactic constraints or rules to the expression of ideas. Being as opposed as possible to ambiguity is the key one, given the aim is a form of language that speaks about the true, definite, certain and absolute.

So unambiguous speech is both a game that can never be won and also the goal to which logicism aspires.

This irritating fact has launched a bazillion forum threads.

Ordinary language use is ambiguous and thrives on that fact. — apokrisis

But why is ordinary language ambiguous? Does that reflect something about the world, our perception of the world, or just the usefulness of ambiguity in natural language?

The snow is white is true iff the snow is white.

But then that depends on what we mean by snow, since snow-like stuff can have different chemical compositions. — Marchesk

So conventionally we mean H2O. But there's nothing prohibiting a different usage. Is there a problem here?

So conventionally we mean H2O. But there's nothing prohibiting a different usage. Is there a problem here? — Andrew M

Maybe not. It's a question of whether deflation needs to take into account meaning.

It's a question of whether deflation needs to take into account meaning. — Marchesk

Only in an abstract sense. The truth schema says "p" is true iff p. You can substitute whatever you like for p as long as it is meaningful.

It does reflect the logic of nature, which is different to the logic of machines.

So nature - as quantum mechanics has confirmed in foundational fashion - is fundamentally spontaneous or indertministic. It is not actually deterministic but simply highly constrained in its habits. Circumstances limit the freedoms.

At least that is my metaphysical argument here. Nature is not a machine. Accident and randomness are part of its inherent reality.

And ordinary language simply follows suit for the same reason - it works. You need the duality of downward acting constraints and then the local freedoms which give the actions to be actually constrained.

So the answer is complex because it both reflects nature at the ontological level. That is the logic of its self organising physicalism (despite our mechanical descriptions of its laws). And also, the brain/mind itself operates with this same natural logic. The brain is not a computer, a machine, but a modelling system seeking to impose informational constraints on the world’s entropic degrees of freedom. The brain is trying to make the world predictable by minimising it’s capacity to surprise.

So language evolved as another level of that regulatory game. It became a collective social medium via which we could impose a regulatory structure on our shared experience and thus minimise the chances of being surprised.

Snow is white. Understood as a constraint on our expectations, we then feel surprise - even alarm - when we encounter a patch of yellow snow. Likewise, we would be surprised if it melted to a gas rather than drinkable water.

So ordinary language is set up with an organic or natural logic. The world is always going to be full of surprises. One can’t know everything, especially when nature itself is inherently indeterministic. A stressed beam is going to surely buckle, thin ice is definitely going to break. But exactly how and when is chaotically unpredictable. That is just way the world is.

Ordinary language builds that fact in. It doesn’t rely on an artificial exactitude. It only has to constrain a state of belief to the degree that something is expected to be more or less the case. Then being wrong is usefully informative, not some disaster. The model of the world can be tweaked by either adding further constraints, or removing existing ones, to improve future performance.

Snow is white, except where the huskies pee. Snow is frozen water, on earth at least.

So ambiguity exists in nature and everyday language is functional because it models nature with that ambiguity in mind.

Logicism is then the application of a purely mechanical notion of causality to the world. It is the language you would speak if the world had the causality of a deterministic machine.

Of course, a mechanised view of nature has been terrifically useful in recent human history. The idea of absolute constraint is a powerful technological vision to impose on the world. There is a reason why we want to treat it as the “true” metaphysics.

But in the end, nature is not actually mechanical. That is just a Platonic vision folk have found useful to impose on its inherent ambiguities.

every snowflake on Mars is unique — Marchesk

I hope we don't run into too many Martians who think they are novel snowflakes.

Yeah, they have a tendency to invade upon being the slightest bit provoked.

a. Extend the pre-existing word-concept to cover all versions of the manifest image and restrict the use of differential vocabulary to distinguishing between the two scientific images which underlie the manifest image.
b. Add a new word-concept in order to distinguish the manifest image by variations in the underlying scientific images.

There is the manifest difference in temperature ranges often found on Earth where C02 ice sublimates into a gas, while H20 turns into a liquid.

Why do you use the word "image" for the scientific understanding? I get the manifest image, because it captures the notion of appearance for humans, given how visually dominant most of us are. But science is more abstract. Is "image" another word for description or model?

If anything option a would be more abstract in that it takes snow to mean any frosted, fluffy white powdery substance that melts in presence of heat. It's a 'descriptive'/'functional' as opposed to a reductive, concrete, substansive definition.

You can figure the sense of a term in context of the conversation and situation, there's no context independent answer. The conversation between that astronaut and the other seems informal and so doesn't require rigid scientific formalism, it's presumably looser and more associative so I want to say the statement is true and he's using snow in a descriptive or functional sense not in a concrete, scientific sense.

So is there supposed to be an essence of snowness, had by, and only by, all snows? And the discussion here is to try to work out what that snow-essence is?

Or are you all just disagreeing about how to use the word "snow'?

I say the latter.

In which case, the discussion is moot.

So is there supposed to be an essence of snowness, had by, and only by, all snows? And the discussion here is to try to work out what that snow-essence is?

Or are you all just disagreeing about how to use the word "snow'? — Banno

Both. We want to be able to use the word "snow" in accordance with something that has snow-like properties. We might limit it by including the chemical composition found on Earth, or we might not.

"X" is Y is true if we define "X" as Y.

Seems obvious.

Have I misunderstood the OP?

Have I misunderstood the OP? — Hanover

I think Apo's post captures my intuition that language can't be reduced to something as obvious as defining "X" as Y in order to arrive at truth.

Both. We want to be able to use the word "snow" in accordance with something that has snow-like properties. We might limit it by including the chemical composition found on Earth, or we might not. — Marchesk

And if we limit it, someone else can change it, limit it in a different way, extend it to mean something else...

It's word games either way.

And we could do the word game with "white" too. Take the scientific definition of white:

"White is the lightest color and is achromatic (having no hue), because it fully reflects and scatters all the visible wavelengths of light".

https://en.wikipedia.org/wiki/White

Now if snow is defined as being white and substance X is not white, it's not snow. But as a lot of snow is not quite white, we have to redifine white or snow is not snow even on earth!

Yes, indeed; and with the word "is", from the discussions of identity hereabouts.

The snow on Venus appears to be lead sulphide. Using language in this way allows for great flexibility.

That's not a problem; indeed, it might be best seen as a strength.

Agree. Definitions are kind of like snow anyway, or snowflakes at least, which we define as distinct entities, but when we try to get our hands on them they tend to melt into each other.