Making books in Low-Tech & LT:Companions 1 & 3

[Turhan's Bey Company]

Originally Posted by Jussi Kenkkilä

What values did you place your calculations on? (words/page, pages/book, pages/pound..)

At the core is a measurement of words per area; w/m^2 to be precise, though any convenient units can be used there. The texts I used in the analysis were all of known length. They were all editions of one of the few standard late antique/Medieval/Renaissance texts: the Latin vulgate Bible or subsets such as the Gospels. Those texts are still widely available and easily counted. The full Bible, for example, is about 712k words (it's certainly possible that any given text might miss some words here or contain a few more there; taking a single word count for any given text was a simplifying assumption). Knowing the word count, dimensions, and number of pages, it's trivial to figure out words per unit of page area. The conversion from that to weight, which relies on exceptionally variable properties of page density and thickness, admittedly has very large error bars attached, but is close enough, as they say, for government work.

[Jussi Kenkkilä]

Using "basic library" as the unit of information, I calculated the costs for different TLs to see what the base price of $3500 consists of (and when was it closest to that value).

The basic library is 100 modern (TL8) books with 500 pages each, totaling 25M words. I'm ignoring the weight and cost of binding and covers for now.

I'm calculating from the basics given in HT and LT, with 0.55 of the monthly pay at average status as the labor production of printers and scribes (Skill 12 professionals).

Printing at TL8 the materials (paper) cost $100, capital depreciation (at 10%/annum) is $6 and labor (offset press) $36, totaling at $142 (4%).

Printing at TL6 the materials (paper) cost $125, capital depreciation is $13 and labor (offset press) $44, totaling at $182 (5%).

Printing at TL5 the materials (paper) cost $1500, capital depreciation is $3 and labor (10 men operating a steam rotary press) $126, totaling at $1629 (47%).

Printing at TL4 the materials (paper) cost $750, capital depreciation is $42 and labor (movable type) $1210, totaling at $2002 (57%).

Printing at TL3 the materials (paper) cost $1500, capital depreciation is $2 and labor (block press) $513, totaling at $2015 (58%).

Hand-copying at TL3 the materials (paper) cost $6000 and labor (cursive at 4000 words/day) $96 250, totaling at $102 250 (292% of $35k).

Hand-copying at TL3 the materials (parchment) cost $60 000 and labor (calligraphy at 1250 words/day) $308 000, totaling at $368 000 (1051% of $35k).

So in summary, printing the library is always cheaper than the $3500 value and hand-writing is always more expensive than the $35k value.

In the case of printing this is most likely because I'm ignoring things like IP in this calculation. In medieval copying culture they didn't matter (except when having to pay rent for exemplars).

Making manuscripts is more costly even at the very high rate of 4000 words/page, which is most likely too optimistic for medieval scribes (averaging at 5 pages/day).

As I previously mentioned, another problem is that on contemporary sources on the costs of making books, the cost of labor is considered to be only half of the total, both for manuscripts and early printing.

All of this applies to commercially produced books and libraries, but are of course applicable to situations with free labor (monks, students) to calculate the time required for copying.

[Varyon]

Originally Posted by Jussi Kenkkilä

Printing at TL4 the materials (paper) cost $750 and labor (movable type) $1210, totaling at $1960 (56%).

I'm not going to go through each example, but I should note here that a handscrew press can manage an entire basic library (25,000 pages) in a single man-month. As a skilled scribe makes $800/month, you're looking at a of $1550. Presumably, the other $2k or so is going into binding, upkeep, organization, and so forth.

Originally Posted by Jussi Kenkkilä

Hand-copying at TL3 the materials (paper) cost $6000 and labor (cursive at 4000 words/day) $96 250, totaling at $102 250 (292% of $35k).

As I noted upthread, a scribe is probably producing at a faster rate than this by taking a penalty. A document that is 74% to 91% (depending on tool quality) perfectly legible, with the remainder being a slightly-harder-to-decipher scrawl, is likely acceptable compared to one that's 98% perfectly legible but takes over twice as long. The former seems appropriate for a typical library, the latter would be more likely found in a higher-quality one.

Additionally, note that the prices given in LTC3 are explicitly for TL4, not TL3. I already showed how the prices can work out at TL4. If you're wanting to match the assumptions from LTC3, you'll want to start with that.

Originally Posted by Jussi Kenkkilä

Hand-copying at TL3 the materials (parchment) cost $60 000 and labor (calligraphy at 1250 words/day) $308 000, totaling at $368 000 (1051% of $35k).

Well, yeah, a piece of art is going to be a lot more expensive than a utilitarian book. GURPS prices for swords are horribly off if you're assuming they have gilded blades and gem-studded hilts as well.

...

All in all, keep in mind that the LTC3 prices and weights are explicitly noted as rough averages, not absolute numbers. If you'd prefer to go with different values, feel free to do so.

[Jussi Kenkkilä]

Originally Posted by Varyon

All in all, keep in mind that the LTC3 prices and weights are explicitly noted as rough averages, not absolute numbers. If you'd prefer to go with different values, feel free to do so.

I want to base my numbers on what's known of real book manufacture, and the scribing speed seems to be the biggest difference. I'd like to calibrate the values to real life, preferably with simple multipliers so I can use the values in the books without having to recalculate everything.

Also on the speed calculations, I'd rather use the skill level 12 for an journeyman/professional level at the listed wages (I'm using the cost of living rather than the $800 to account for different TLs). As mentioned in LT:C3, masters will have skill level of 14+, but also one higher status (double wages). A master hurrying with -2 to skill will take 80% of the time of a journeyman, producing only 125% as much at 200% of the cost. This way I'll know the average time and cost and can then take into account changes in skill and circumstances.

[Jussi Kenkkilä]

Originally Posted by Varyon

As I noted upthread, a scribe is probably producing at a faster rate than this by taking a penalty. A document that is 74% to 91% (depending on tool quality) perfectly legible, with the remainder being a slightly-harder-to-decipher scrawl, is likely acceptable compared to one that's 98% perfectly legible but takes over twice as long. The former seems appropriate for a typical library, the latter would be more likely found in a higher-quality one.

Additionally, note that the prices given in LTC3 are explicitly for TL4, not TL3. I already showed how the prices can work out at TL4. If you're wanting to match the assumptions from LTC3, you'll want to start with that.

The lower legibility would be something that is taken into account on the quality modifier for the book price. So unless the assumption is that only fine-quality books are easy to read, I would keep the speed at 98% legible level.

For TL4 movable type printing the ca. $2000 is rather close to the library value given.

For the same TL cursive manuscript the total is $113k. (Actually more expensive than TL 3 if we take into account the increase in labor productivity. If not, the higher price of early paper will make TL3 more expensive than TL4).

If we use the same number of pages for that too (not multiplying by 4 for hand-writing), the scribe-work will cost $27.5k, for a total of $30.5k. As this is close to the value given for hand-written library, maybe the increase in page count was forgotten in the estimation of the price multiplier. So for a cursive manuscript library, we would multiply the prices given by 40 instead of 10.

As for the cursive vs. calligraphy issue, we could think of the cursive as the cheaper, less legible method of book production.

For a daily speed of 32 pages for cursive, 10 for calligraphy and 2 for illumination we could approximate those to quality levels: cursive for "good", calligraphy for "fine" and fully illuminated for "very fine".

Maybe say that "fine" books are fully legible and the "good" rate will only produce (skill 12+6-8=10) 50% legible text in 20% of the time it takes to copy a book with calligraphy. "Cheap" would take the maximum reduction, for 37.5% legible text in 10% of the time. This would mean that the real price of cheap and good libraries would be the extra effort in reading them. This way the quality wouldn't only be about resale value and prestige.

With these quality levels a cheap hand-copied basic library on TL 3 paper would cost $36800, which is close enough, although for pre-paper societies some other cheap material like leaves would be used.

Even with these modifications the high cost of parchment will push the prices of better libraries up. Because paper had a long time stigma in Europe as a poor substitute for parchment, it wasn't used in better books. The choice of materials should be taken into account on calculating the total costs for libraries of different qualities.

[Peter Knutsen]

Originally Posted by Polydamas

Ninety words per page sounds like it might include manuscripts which had more picture than text, or very small manuscripts. Fifty thousand words per volume seems about right.

The Ars Magica RPG consistently insists that books that are not "illuminated" are much less useful than books that are. Given how well researched it tends to be, and how well versed its line editor is in matters medieval, I'm inclined to just accept that position as being fact.

[Jussi Kenkkilä]

Originally Posted by Peter Knutsen

The Ars Magica RPG consistently insists that books that are not "illuminated" are much less useful than books that are. Given how well researched it tends to be, and how well versed its line editor is in matters medieval, I'm inclined to just accept that position as being fact.

This would also mean that the extra cost of illumination affects the utility of the book. So except for gilding and other extravagance, the price of a book should be tied to it's utility. Maybe give it an equipment modifier that affects reading speed, learning from books, research and data search rolls.

This would also nicely keep the value of the book independent of the actual information within. As in manuscript cultures the value comes from the manufacturing, information has almost zero value (once you copy a book you can keep on copying without having to pay for the content).

[Flyndaran]

Great for magical tomes, but not really applicable to reality.
Degree of fanciness rarely coincides with usefulness.
Haven't we all read pretty calligraphy fonts that slow reading? And obviously anything but very tiny illumination is all about pretty, taking up space that could be used for more words.

[Turhan's Bey Company]

Originally Posted by Flyndaran

Degree of fanciness rarely coincides with usefulness.
Haven't we all read pretty calligraphy fonts that slow reading?

That, I strongly suspect, is a question of familiarity rather than fanciness vs. utility. If, say, blackletter and ligatures were inherently difficult to read, I can't imagine that they'd have been used for centuries. Now, modern calligraphic styles may substitute aesthetic effect for utility, but at this point, manuscript traditions are pretty much dead, and those styles are deliberately designed to look good, not to be read. Different bag of cats.

That said, I largely agree that fanciness and utility are orthogonal. A Bible, for example, might gain a bit of utility by larger capitals at the start of each chapter, verse numbers set off to the side, and the words of Christ in a different color of ink, but does it really become easier to read and comprehend if there are jousting muskrats in the capitals and a shiny splash page for every book? I don't think so. Now, they certainly have utility in a social context. Fancier stuff impresses people (and GURPS has rules for that, of course). That's why books like that were produced. But as an improved information resource? I'm not seeing it.

[Peter Knutsen]

Originally Posted by Flyndaran

Great for magical tomes, but not really applicable to reality.
Degree of fanciness rarely coincides with usefulness.
Haven't we all read pretty calligraphy fonts that slow reading? And obviously anything but very tiny illumination is all about pretty, taking up space that could be used for more words.

As far as I known, one aspect of the Carolingan Reform was the introduction and dissemination of a simpler caligraphy, one that was quicker and more efficient both to write and to read. And in general, I don't think medieval books are caligraphed in a way that makes them hard to read, although personally, I find even modern handwriting hard to read. Not just my own which is clearly barely legible, but even that of others. I seem to partly have lost the ability to read handwriting easily, and partly I lack the patience.

But most medieval texts? I'd imagine they were pretty easy, if I knew Latin.