The Crystal Goblet (1955) | Hacker News Reader

The Crystal Goblet (1955)

web.archive.org

28 pointsTomte15 comments

Matthew Butterick disagrees strongly with the Crystal Goblet[1]. His argument that it's impossible not to communicate with typographical choices convinced me to doubt the contents of the text “The Crystal Goblet”.

[1] https://practicaltypography.com/drowning-the-crystal-goblet....

I do everything I can to make typography invisible to me. I do this by converting it all to the same style. I don't allow websites to chose their own fonts. I use Firefox's Reader View whenever possible. I remove all formatting from ebooks and apply my own CSS in Calibre. I'm convinced that the most legible typography is the one you're most familiar with, and if you consistently read the same typography it will become the metaphorical crystal. The actual design of the type matters very little.

I might do that myself, that sounds really cool. I've never imagined doing that before.

Butterick is claiming a significance for the appearance of text that simply is not there. Are an author's thoughts incomplete until they have been typeset and printed?

That's a question that misses the thrust of the argument - how do you intend to read an author's thoughts?

Butterick is discussing text, and text doesn't exist without typography (or handwriting).

> Butterick is discussing text, and text doesn't exist without typography (or handwriting).

That argument is essentially the empty tautology "you can't have text without writing." I don't agree with him, but I am not claiming that his argument is anything like that trite.

My opinion is that Butterick gives way too much significance to presentation. He writes "presentation matters specifically because it’s not meaningless", but it is extremely rare for it to significantly add to the meaning.

>but it is extremely rare for it to significantly add to the meaning

Then what exactly are you disagreeing with him about? Your initial comment suggested to me that you thought typography had no significance to text. Now you say that it does - but Butterick doesn't say something about how important typography is, only that it is important. Just that simpler claim is his objection to the Crystal Goblet.

So what sort of importance do you think Butterick thinks typography has? Minimally important, perhaps? Trivially important? Insignificantly important? Unimportantly important?

What you are trying here is a weak-man fallacy, restating Butterick's objection in the weakest possible terms, in the hope that there's no gap between it and the dissenting opinion for the dissenter to get any leverage. It's like a boxer's clinch, holding on tight to his opponent so that his opponent cannot take a swing at him.

Unfortunately for you, "importance" just isn't suitable for this ploy, because "unimportantly important" just means not important. It is nonsense to suggest that Betterick did something so pointless as to write this article because he differs in the slightest possible way from the position set out in the Crystal Goblet. Of course he thinks there is a significant difference.

Maybe if it were typeset differently, you would see that?

>So what sort of importance do you think Butterick thinks typography has? Minimally important, perhaps? Trivially important? Insignificantly important? Unimportantly important?

I think to discuss that is beside the point - the article is not making a relative argument. Butterick claims that it is important.

>It is nonsense to suggest that Betterick did something so pointless as to write this article because he differs in the slightest possible way from the position set out in the Crystal Goblet.

Why? The Crystal Goblet puts forward an absolutist position - typography is so subservient to writing that it is meant to be "invisible". The gradients of "how much" typography matters are all significantly more similar than how they all differ to claiming that typography doesn't matter at all.

I don't think Butterick really feels that typography matters the minimal possible non-zero amount - but even if he did, he would already be miles away from agreeing with the Goblet, because that would already be a completely different approach to text. Besides which, Drowning the Crystal Goblet makes no claim on "how much" - so objecting to the significance he places on typography requires either

  * assuming how much significance he places on it, and then objecting to that, or
  * objecting to any amount of significance

Firstly, you are mistaken about the Crystal Goblet, and maybe, in this case, Butterick is doing the same. It is predicated on the assumption that typography does have an effect; Warde goes on to claim that it is at its best when it is not ostentatiously drawing attention to itself.

Secondly, you are attempting to create a straw man by insisting that the Crystal Goblet must be (mis)read as an absolute manifesto, but if you need to do that, you don't really have much of an argument against it (I don't think Butterick is making this error.)

Thirdly, you are waffling over whether Butterick is claiming a significant difference between himself and Warde. By the fifth paragraph, you are tentatively acknowledging that Butterick is (probably) making such a claim, which is wise, because otherwise your argument would quite clearly just be pointlessly and tiresomely pedantic (something that I would not accuse Butterick of being.) Immediately after that, however, you resume being pedantic by bringing back the straw man, saying that it would not matter if Butterick was not claiming a significant difference because (your misunderstanding of) Warde's position is that it is an absolutist one. This sort of having-your-cake-and-eating-it sometimes goes by the name of a motte-and-bailey fallacy.

Maybe we should just go back to my original point: what, of significance, is a typical work of prose missing before it is typeset? you might actually be able to make a case for mathematics, some poetry and possibly medieval illuminated manuscripts, but these are corner cases in comparison to the global corpus of prose.

>what, of significance, is a typical work of prose missing before it is typeset?

The ability to be read.

So Butterick's article can be reduced to one point: "The Crystal Goblet is wrong because you can't read an article until it is typeset." I may be mistaken, but I think you have misunderstood Butterick's response even more than you misunderstood the Crystal Goblet itself.

The larger discussion that this connects to is about the relationship between form and content. Generally, people you might describe as sophisticated critics tend to dismiss the idea that form can be a neutral container for content. Form and content are "indissoluble" or they should form a "unity", or at least be in "harmony". Often the goal is for the form to be in the service of the content, subtly supporting and amplifying it.

In practice, tailoring form to content, or pursuing form for its own sake, requires more resources. It's a luxury. Standard formats are the most economical way to transmit text, for the producer. But throwing away the skill of the typographer altogether (just using standard forms/formats) means throwing away the artistic experience that typographic design can sometimes achieve, and the reader loses out in that scenario.

It's routine to hear cinematographers and film editors talk about how their craft should be invisible. That doesn't mean they force every scene into the same straightjacket, or that you can't see their craft if you look.

Typography is a little bit different, but it is often obvious when a typographer is trying hard to show off. Whether that kind of result appeals to you or not is a matter of taste.

When designing HyperStudio in 1988 and beyond, I had a similar philosophy, that the software tool should strive to be invisible to the tool. There was an occasion when we asked a user how they thought the interface could be improved, and they responded "what interface?". I don't presume that we were exceptional in achieving our goal, but we did try, and this article is an interesting support of that goal. I'm also reminded of what I read about Japanese art, what are the fewest lines needed to create the image and story?

Ironically, because of the very small font and the wide paragraphs (on my phone; Firefox on Android), I found reading this to be very difficult, as I had to scroll sideways a lot after I zoomed to the point where I could read it.

Reply on news.ycombinator.com