Is the invented script of Luigi Serafini's "Codex Seraphinianus" a real (or enciphered) language, and how...

Jim Gillogly wrote:
> The numbering system appears to be real, though.  I don't remember the
> details, but I think it's base 21 or 22 with some little fillips
> thrown in with the higher digits.  Sometimes.  But the page numbering
> appears to be consistent with that hypothesis.

Base 21; the ones digit goes through a repeating sequence of 21 glyphs,
and the higher places are represented by an (unrelated) sequence of
strokes and vees, with the ones digit inserted in one or more fixed
places.  The page numbering follows that sequence, except for a couple
of "extra" pages which have variant numbers, presumably representing
"17 and a half" or "17.a" or something.  But there isn't enough
information to work that out.

I never got anywhere with the language.  Remember, however, that if
it's an encryption of a real text the plaintext is likely to be
Italian....

-- Andrew C. Plotkin <ap1i+@andrew.cmu.edu>  (April 1994)

I don't think so: the guy whose right arm is a fountain pen has been
working in French.  Too bad the Rosetta Stone is in Seraphianese and
something else we don't know.

Not English, though.

-- Jim Gillogly <jim@rand.org>  (April 1994)

Those interested in the Codex Seraphinianus might wish to look up the
previous discussion of this that took place on the SF-Lovers list back
in May - July 1987.

-- William G. Martin <ar710@cleveland.Freenet.Edu>  (April 1994)

[ Editor: Martin then reposted that discussion, introducing it: ]

Here is the 1987 discussion of CODEX SERAPHINIANUS from the SF-Lovers
mailing list; I found it on a diskette where I had saved it.

I think those of you who have recently stated that it can now only be
found at prices like $250 - $350 will be particularly galled to
discover that it was being sold remaindered at $25 or so...

Note this is a 7-year-old discussion, so probably every e-mail address
in this is no longer valid.

[ Editor: the 1987 SF-Lovers messages follow, in their original order. ]

Well, I received my long-backordered copy of CODEX SERAPHINIANUS from
Publishers Central Bureau yesterday.  Now, I know that at least one
other person out there has a copy of this -- it was a reference to this
book on SF-Lovers that caused me to notice it when it showed up in one
of the PCB catalogs -- and I have some queries that I hope someone out
there can answer about this.

(For the uninitiated, this book is a rather strange art work.  It is a
several-hundred-page large-format book, entirely hand drawn in the
original (that is, no printed text, but all pages are drawn), mixing
colored illustrations with pages of text and captions, entirely in an
imaginary language, describing a completely alien world (except that
the inhabitants appear to be human in form).  An extremely surreal
exercise, the book is a sort of encyclopedia describing this world and
its cultures.)

Author: Luigi Serafini Copyright 1983 Abbeville Press [I think -- I'm
doing this from memory...]

Now, my queries:

1) I understand that the alien language in which this book is written
is actually a completely-worked-out real language; that is, the
mysterious squiggles are not just random marks, but could be translated
to real text.  Does anyone know if this is really another language, or
just a consistent transliteration of some ordinary language, and, if
so, what that language is?  (I suspect it would be French; there is one
drawing of a manlike creature with a pen as its arm creating a page of
text which is French in script -- I suspect that this is a
self-portrait of the artist, doing in the book's society what he was
doing in reality [that word gets a bit variable in this context! :-)];
creating an alien culture.  So I wonder if the text might be really
French in a simple substitution cipher -- this funny squiggle means
"a", that one means "b", and so on -- as opposed to being an actual
different language.)

2) In any case, is a translation into English available anywhere?  I
realize that the book can be viewed as a wonderful puzzle, and figuring
out the meaning(s) is a lot of fun, but I neither have the time nor am
I up to the effort required.  I'd like to cheat and buy a crib sheet...

3) My copy came with the inclusion of an oddly-cut fragment of
photograph, perhaps cut from a calendar, with the number "031" written
on the back, stuck between the pages.  Is this actually the copy
number, put in by the artist, indicating that this is copy 31 of <some
number>?  Or is it just a scrap of paper with no meaning?  (Or is it a
message from an alien power so subtle and mysterious I cannot interpret
it?....  :-)

[If it is the real copy number, that certainly is odd.  Why would such
a low number as "031" be remaindered?  I don't know what this cost when
new, but its weird enough to have sold at least several hundred copies
no matter how much it was, just to the esoteric artsy types...]

4) Can anyone point me to any reviews or commentaries or other sources
of information or discussions about this book or the artist?

-- Will Martin <wmartin@almsa-1.arpa>  (12 May 1987)

A friend has done considerable research on the book, and has found that
is is definitely NOT a simple transposition cipher on any known
language.  No matter what he tried he could not get the letter ratios,
specifically the "vowel"/ "consonant" ratios to come out consistently
and correctly.  Also, he managed to decipher the page numbering scheme,
and reports that it is in a very odd prime-number variable base scheme
(first level has seven symbols, second level 13, third 23, or something
like that.)

>2) In any case, is a translation into English available anywhere?

Not that any of us have ever found (I have about five friends who LOVE
the book, including myself)

>4) Can anyone point me to any reviews or commentaries or other
>sources of information or discussions about this book or the artist?

Douglas Hofstadter [Editor: "Hofstadtler" in the original] (who wrote
Godel, Escher, Bach - a book I also highly recommend) talked about the
Codex in one of his Scientific American columns "Metamagical Themas".
That's how I was turned on to it.

-- Mike Farren <hoptoad!farren>  (13 May 1987)

The Publishers Central Bureau stock # for Codex Seraphinianus is 475146.
The price is $24.95 (original price was $75.00).  (By the way, the
original copyright date was 1981, and the Abbeville Press edition is
dated 1983.)

Publishers Central Bureau
One Champion Avenue
Avenel, New Jersey 07131

I just spent several lunchtimes in the St. Louis Public Library looking
for more info on this book.  They have the Codex itself in the art
department, but I could find no reference to reviews of the book, or
articles about it or the author, in any of the standard references or
periodical indices, except a pointer in the *1984* (not 1983) New York
Times Index to what turned out to be a totally worthless tiny
one-paragraph "review" (really just a short description of the book).
[In case anyone is interested, to save you the effort, that is Feb 12,
1984, Section 7, Page 20, Col. 2.]  I am amazed that a reference like
the "Art Index", for example, has nothing about the book nor the author
(except a pointer to an article in "Domus" that refers to some other
unrelated works by him).  How could such an artwork go so unremarked?
Is it that it is so strange and outre' that critics cannot think of
anything to say about it and therefore ignore it?

I also spent some time searching bound volumes and microfilm of
Scientific American, and just cannot find the Hofstadter [Editor:
"Hofstadtler" in the original] "Metamagical Themas" column that
discusses this book.  Either I just plain missed it, or the book is not
referred to in the title of the column, nor in the bibliography page at
the rear of each issue (which does contain references for other books
mentioned in Metamagical Themas).  I did scan most of the text in any
column which seemed appropriate.  Maybe this was another one of those
two-sentence references buried in amidst much other prose, like the NYT
"review", which will turn out not to be worth the effort of unearthing
it?  I searched 1981 through the last of these columns, which terminated
in late '83, without finding the citation.

Anyway, I guess the Codex will have to be appreciated on its own,
without any guidance from published criticism or other "authoritative"
commentary.  It is still somewhat surprising that such an astounding
book seems to have sunk from view without a ripple of discussion...

-- Will Martin  (18 and 27 May 1987)

[ Editor: Martin's message of 18 May, with the same stock number and
library search, is merged into his fuller message of 27 May above. ]

>I also spent some time searching bound volumes and microfilm of
>Scientific American, and just cannot find the Hofstadter
>"Metamagical Themas" column that discusses this book.

You couldn't find the reference in Scientific American because there is
no reference to CS in any of Hofstadter's columns.  The reference occurs
in the *book* Metamagical Themas: Questing for the Essence of Mind and
Pattern (Douglas R. Hofstadter, ISBN 0-553-34279-7) which is a
collection of all Hofstadter's SA columns and some other material, with
commentary and additional new material added to each for the book.  The
reference to CS is in a 'postscript' added to the end of the reprinted
December 1982 column entitled "Stuff and Nonsense" on p. 229 of the
large format softcover edition.

-- William H. Ansley <ansley@buffalo.csnet>  (28 May 1987)

It is indeed amazing that the book has been remaindered without
generating any discussion...I'm especially surprised that NO ONE has
reviewed it in an SF magazine (unless it was reviewed in some obscure
fanzine).

-- William Ingogly <wfi@rti.rti.org>  (27 May 1987)

According to an online book review database, the CS was reviewed in
several places, although I cannot vouch for the quality of the reviews.

1) Esquire Magazine, V 102 p260, Nov. 1984
2) NYT Book Review, V 89 p20, Feb. 12, 1984 (I think this was already
   mentioned as being a short blurb, not a real review.)
3) Los Angeles Times Book Review, p4, Dec. 18, 1983
4) Choice, V21 p696, Jan. 1984
5) Atlantic Monthly, V 253 p105, Feb. 1984
6) Publishers Weekly, V224 p62, Oct. 28, 1983.

-- Aline Baeck <NORRIS@STRIPE.SRI.COM>, SRI International  (3 June 1987)

My thanks to Aline Baeck for locating the references to reviews of the
Codex Seraphinianus!  I just spent some time at the St. Louis Public
Library looking them up, and thought I'd pass the word.

Here are the citations she provided, along with comments I noted when
reading the reviews.  Notice that all of these that I found are very
short, one paragraph or so in length.  They are mostly in the nature of
blurbs of praise or brief descriptions, rather than formal reviews.
They could all have been summed up as "Gee whiz -- what a weird strange
book!  How neat!"-- not really terribly valuable from a critical
standpoint...  (They were all in microfilm form, by the way.)

> 1) Esquire Magazine, V 102 p260, Nov. 1984
Describes the CS as the "oddest book of the year" and as an
"untranslatable" description of a "parallel universe".

> 2) NYT Book Review, V 89 p20, Feb. 12, 1984
Mentioned in an earlier posting of mine, this is a brief description
with a small reproduction of one of the book's pictures.

> 3) Los Angeles Times Book Review, p4, Dec. 18, 1983
Unfortunately, this one I can't find; the library gets this paper but
doesn't keep back issues on microfilm.

> 4) Choice, V21 p696, Jan. 1984
This is actually a periodical named "Choice Books for College
Libraries" or something like that.  Describes the CS as "delightful".

> 5) Atlantic Monthly, V 253 p105, Feb. 1984
Refers to the CS as "completely alien".

> 6) Publishers Weekly, V224 p62, Oct. 28, 1983.
Describes the CS as "haunting and a source of endless stimulation".

If anyone happens to locate the LA Times review, let me know if that is
just another one of these same sort of blurbs.

(I'm still waiting for the book version of "Metamagical Themas" to
return to the library shelf so I can read the reference to CS therein.)

-- Will Martin  (10 June 1987)

In case you haven't yet found Metamagical Themas, here are the
paragraphs about the Codex Seraphinianus (from p. 229 of MT):

   Codex Seraphinianus is a much more elaborate work.  In fact, it is a
   highly idiosyncratic magnum opus by an Italian architect indulging
   his sense of fancy to the hilt.  It consists of two volumes in a
   completely invented language (including the numbering system, which
   is itself rather esoteric), penned entirely by the author,
   accompanied by thousands of beautifully drawn color pictures of the
   most fantastic scenes, machines, beasts, feasts, and so on.  It
   purports to be a vast encyclopedia of a hypothetical land somewhat
   like the earth, with many creatures resembling people to various
   degrees, but many creatures of unheard-of bizarreness promenading
   throughout the countryside.  Serafini has sections on physics,
   chemistry, mineralogy (including many drawings of elaborate gems),
   geography, botany, zoology, sociology, linguistics, technology,
   architecture, sports (of all sorts), clothing, and so on.  The
   pictures have their own internal logic, but to our eyes they are
   filled with utter non sequiturs.
      A typical example depicts an automobile chassis covered with some
   huge piece of what appears to be melting gum in the shape of a small
   mountain range.  All over the gum are small insects, and the wheels
   of the "car" appear to have melted as well.  The explanation is all
   there for anyone to read, if only they can decipher Serafinian.
   Unfortunately, no one knows that language.  Fortunately, on another
   page there is one picture of a scholar standing by what is
   apparently a Rosetta Stone.  Unfortunately, the only language on it,
   besides Serafinian itself, is an unknown kind of hieroglyphics.
   Thus the stone is of no help unless you already know Serafinian.  Oh
   well...  many of the pictures are grotesque and disturbing, but
   others are extremely beautiful and visionary.  The inventiveness
   that it took to come up with all these conceptions of a hypothetical
   land is staggering.
      Some people with whom I have shared this book find it frightening
   or disturbing in some way.  It seems to them to glorify entropy,
   chaos, and incomprehensibility.  There is very little to fasten
   onto; everything shifts, shimmers, slips.

-- James J. Lippard <Lippard@MIT-MULTICS.ARPA>  (17 June 1987)

Described as by "Luigi Serafini", but the Copyright is held by France
Maria Ricci.  This is not a review, but an opinion; and I assume you
know what this book is.

Well, I borrowed it from the local library.  The drawings are
interesting, but the "text" is, in my opinion, a fake.

(a) the page numbering scheme makes no sense

(b) I can't find any obvious regularities in the words.  I looked for
    inflection at the ends and in the middle of words, for special
    initial letters, and for prefixes.  (You know, the tricks Ventris
    applied to Linear B).  No luck.

(c) many sections of text seem to be repetitions of the same symbol
    groups, over and again, with trivial variations.  Nothing like a
    sentence structure.  Moreover, the "letter" distribution is not
    consistent between pages.  This is scribble, not script.

Does anyone have, or recall, any reason for believing the text IS
language, real or invented?

-- Robert Firth <firth@sei.cmu.edu>  (20 July 1987)

[ Editor: the copyright holder is given as "Franco Maria Ricci" in the
Library of Congress data quoted in cryptology/Voynich. ]

I've been trying to decode the numbering in Luigi Serafini's Codex
Seraphinianus, and have a couple of questions that other decoders might
have answers to:

1) any ideas as to why the prefix switches from "|" to "V" from number
56 to 57, and from "V|" to "<inverse-V>V" from number 77 to 78, and at
similar 56+n*21 (4>=n>=0) boundaries?  After all, there is no such
switch from 14 to 15, or from 35 to 36, or from 161 to 162???  It is
curious that this occurs at (some of the) multiples of 7 boundaries,
while the number system is obviously base 21.

2) why is number 169 (=13 squared, curiously enough) represented so
inconsistently?!  (Instead of being the expected
"||<symbol-for-1><symbol-for-1><symbol-for-1>|" it is
"||<symbol-for-2><symbol-for-0><symbol-for-3>|").  Following this, the
"usual" base-21 numbering continues, but is now thrown off by one due
to this glitch (and it doesn't seem to be a spurious page, since the
same effect occurs later in the book when the page numbering has
restarted from zero).

-- Martin S. Feather <feather@vaxa.isi.edu>  (20 July 1987)

This message describes the page-numbering scheme of the Codex.  It
answers some questions other SF-lovers have asked about this particular
aspect of the Codex.  It summarizes all we know, and poses some
unanswered questions.

The Codex is in two (originally physically separate) volumes, which I
call vI and vII.  There are eleven major sections, five in vI and six
in vII.  We can guess at the topics of each of the sections, although
some are problematical.

vI: Plants
    Animals
    Nonhuman Bipeds
    Physical Science
    Technology

vII:Humans
    History
    Writing
    Food and Clothing
    Games
    Architecture and Urban Planning

The eleven sections each begin with a right-hand page bearing a
centered title in capitals.  The section title pages also bear a "topic
illustration" excerpted from the section proper.  All the "internal
sections" (that is, all the sections except Plants and Humans) have a
facing blank left-hand page.

Most of the pages of the Codex have centered sequence marks on their
bottom edges.  Exception: most of the section title pages and their
facing pages lack sequence marks.  Exception to the exception: the
section title pages for Humans and Games do bear sequence marks.

The two volumes use the same sequence marks.  That is, the system
starts over again at the beginning of vII.  Every single sequence mark
that is missing from vI (due to pages that don't bear sequence marks)
is supplied in the appropriate place in vII.  vI and vII are completely
consistent with each other (allowing for moderate scribal variation),
so we have a fair degree of confidence that there are no "mistakes" in
the sequence as far as we have it.

The sequence marks fall into groups of 21, which I call "scores".  I
give each score a number starting at s0.  The two volumes end at
slightly different places, both in s8.

The scores are built on a (nearly) invariant system of 21 "digits".  To
specify a page, I put a period between score and digit number.  For
example, the Death of the Heretic Scribe is in Humans, II:1.14-15.

Each score falls into two sections, the early part (digits 0-14) and
the late part (digits 15-20).  To derive the sequence mark for a given
page, you (usually) substitute the digit (in the places marked by x's
in the following table) into a template whose fixed features are
vertical strokes (which I write "I"), V's, and upside-down V's (which I
write "A").  Each score has two such templates, one for the early part
and one for the late part.  But in s0, s1, s7, and s8, the early and
late templates are identical.  The templates are:

Score   Early Template  Late Template
0       X               X
1       IX              IX
2       IIX             VIX
3       VIIX            AVIX
4       AVIIX           IAVIX
5       AVIIXX          IAVIXX
6       XIIXX           IXIXX
7       IXXXI           IXXXI
8       IIXXXI          IIXXXI

There are four additional complexities that currently have no good
explanations.

First, two of the digits (1 and 15) look almost identical, and I cannot
tell which differences are significant and which are due to scribal
variation.  Thankfully, 1 is an early digit and 15 is a late one, so
for those scores with differing early and late templates, the template
provides the information that the digit doesn't.

Second, two of the digits (5 and 7) are based on the same A-shape that
is used in the score templates.  Their only apparent difference is that
7 has a trailing dot that 5 lacks.  When 5 and 7 are juxtaposed with
themselves or with template symbols, they undergo bewildering
transformations.  Dots appear and disappear, V's and A's turn
upside-down, and so on.  I suspect this is all due to "contact
constraints" but I can't formulate them.

Third, two of the digits (9 and 19) are drawn differently in different
scores.  19 gets an extra bump in s6, and 9 gets one in s1, s6, and s7.
These variations are consistent between the two volumes.

Fourth and finally, there is the matter of the unfortunately incomplete
s8.  What we have of it looks as if it has been hit with hard gamma
radiation.  It shows three serious mutations:

1. An additional sequence mark ("8.05") has been inserted between 8.0
and 8.1.  It looks like "II 203 I", and is the only "heterodigitous"
mark in the whole page sequence.

2. 8.4 is missing.

3. 8.9 is missing.

All these peculiarities are maintained with complete consistency
between vI and vII.

If our moderator deems this article to be of sufficient interest to
actually include it in a digest, I will make another posting in the
near future, arguing that the language itself is spurious, and devoid
of actual linguistic content.  In addition I will try to explain some
design principles that run through this remarkable work.

-- Allan C. Wechsler <ACW@waikato.s4cc.symbolics.com>  (23 July 1987)

I am _extremely_ interested in obtaining a copy of the afore-mentioned
book, Codex Seraphinianus by Luigi Serafini.  I have read with great
interest all of the postings, but asking my mother (who is a
pseudo-librarian) about it got a flat zip.  Apparently this book is not
in print at moment, and all of the book stores I have tried don't have
it.  I also heard that there exists two copies of this book, an
expensive hardbound version and a two-volume soft copy.

-- Matt <mdharding@watmath.waterloo.edu>  (31 July 1987)

Publishers Central Bureau is no longer a source for 'Codex
Seraphinianus'.  They returned my order with a notation that "...we can
no longer supply or cannot recognize the item."

-- Rodney Hoffman <Hoffman.es@Xerox.COM>  (28 July 1987)

The same thing happened to me.  Does ANYONE have any alternate source
for this book?

-- Bill Ingogly  (29 July 1987)

[I got my copy from

   Edward R. Hamilton, Bookseller
   Falls Village, CT  06031-5000

 but I didn't notice it in the last catalog (I just skimmed it though).
 Write him for a catalog or specifically ask him.  Hope this helps.
 -Dave]

[ Editor: the bracketed note is by the SF-Lovers digest moderator. ]

As of 3 months ago Soap Plant gift/novelty/boutique shop in West
Hollywood, CA., had a "large supply" of "Codex" (according to them).
The price was around $40+.  If there is interest, I could find out more
(it's a bit out of my usual routes)...

-- Oleg Kiselev <oleg@quad1.quad.com>  (31 July 1987)

Yes, yes I'm interested.  An address or phone number will do me just
fine.  Do you know if the Soap Plant had the hardback or softback
edition?

-- Lauri Howard <howard@ames-pioneer.arpa>  (3 August 1987)

Oleg, if you'd post this information it would be *greatly* appreciated
since other sources for the Codex seem to have dried up.

-- Bill Ingogly  (3 August 1987)

:As of 3 months ago Soap Plant gift/novelty/boutique shop in West
:Hollywood, CA., had a "large supply" of "Codex" (according to them).

Alas!  No more.

-- jeff putnam <putnam@ge-crd.com>  (5 August 1987)

The Moderator sent me an enthusiastic reply to my last long screed
about the page-numbering system in the Codex.  Flattery will get you
everywhere.  Here is why I believe the "language" of the Codex is an
elaborate fraud, devoid of actual semantic content.

It's very hard to isolate letters from the cursive script, so I
concentrated my efforts on the "upper-case" or "heading" alphabet.

I took a very small sample of text in the "heading" alphabet, to wit,
the eleven section headings.  I did a very naive statistical test.  I
wanted to answer the question: do the headings behave statistically
like samples from the same language?  In summary, they do not.

In fact, the frequency distribution is suspiciously flat, as if the
artist were trying to be "fair" to each of his invented letters.
Furthermore, certain letters come into vogue, that is, become very
frequent, for a couple of headings, and then disappear again.

It's hard to think of a reason why a real language would show such a
distribution.  The most obvious theory is that those letters came into
vogue as they were invented.  The artist was successively infatuated
with each of his new letterforms, but then got tired of each.

So the naive statistical evidence says that the headings are
nonlinguistic, and it would be darned surprising if the main text made
sense while the headings didn't.  Hence, my current belief is that none
of it makes sense.

By the way, if people are interested in this sort of thing, check out
"Parallel Botany" by Leo Lionni (sp?).

-- Allan C. Wechsler  (24 August 1987)

ACW@waikato.s4cc.symbolics.com writes:
>...  Here is why I believe the "language" of the Codex is an
>elaborate fraud, devoid of actual semantic content...

The Mayan Codex was an elaborate and beautiful artifact whose semantic
content was for many years a complete mystery.  Yet people reacted to
that artifact as an object of beauty in spite of/because of the
mystery.  Similarly, Egyptian heiroglyphs were for many years objects
of beauty as well as semantic mysteries.  Maybe Luigi Serafini (if I
have his name right) was trying to do something similar: create an
undecipherable text whose SEMANTIC content is (perhaps) nil but whose
AESTHETIC content is considerable ... and maybe the controversy
surrounding the "deciphering" of Serafini's text is part of the ART in
what he's done: art as process, in a certain sense: the controversy and
ambiguity is in fact part of the aesthetic experience of the Codex.

-- Bill Ingogly  (25 August 1987)