The identity
- Code point
U+237C- Unicode name
- RIGHT ANGLE WITH DOWNWARDS ZIGZAG ARROW
- Block
- Miscellaneous Technical
- HTML entity
⍼- In Unicode since
- Version 3.2, March 2002
- Known meaning
- None
- Known usage
- None
The name is a contraction. Its older SGML entity name was
angzarr, from the description angle with down zig-zag
arrow — and that description is itself the only definition anyone has
ever supplied. It says what the glyph looks like. It does not say what it
means. Every subsequent standard copied the shape and inherited the silence.
Chain of custody
The glyph was never invented by a standards body. It was inherited, one document at a time, each editor reasonably assuming the previous one knew why it was there.
-
1949 – 1960s
H. Berthold AG type catalogs
The German foundry lists the glyph under Mathematical Symbols with a gloss: Azimut, Richtungswinkel — azimuth, direction angle. This is the earliest interpretation anyone has found, and the last one ever offered.
-
1954
Absent from Monotype
The Monotype catalog of this year does not contain it.
-
1963
Monotype catalog — matrix serial S9576
It appears, filed under Special Signs: Arrows and Shafts. No explanation accompanies it. Someone at Monotype cut a physical matrix for this shape, which means someone, somewhere, had ordered type set with it.
-
1964
Herman Holmqvist, Sweden
A Swedish typographic handbook describes it as riktningsvinkel — direction angle. The azimuth reading, repeated. Still no mathematical text using it.
-
1972
Monotype mathematical characters — reassigned S16139
The same glyph, a new serial number, in a catalog specifically of mathematical sorts. Why it was renumbered is not recorded.
-
15 October 1986
ISO 8879 (SGML) — without it
The SGML standard ships its ISO AMS-A entity set. The glyph is not in it. This matters: it dates the entry point to somewhere in the following five years.
-
1 July 1991
ISO/IEC TR 9573-13 — first verifiable appearance
The glyph is now in ISO AMS-A, as entity
angzarr, described angle with down zig-zag arrow, carrying AFII glyph code D97C (decimal 55676). Edited by Anders Berglund. It is, in effect, laundered: a foundry's undocumented type sort is now a character in an international standard. -
1993
AFII International Glyph Register
Entry 55676 / D97C: “right angle with down zig zag arrow.” The description is copied forward again, verbatim in substance. No source is cited, because there is none to cite.
-
24 June 1997
STIX — Barbara Beeton collates the tables
The scientific and technical publishers' font project draws its character list from the ISO technical report. Beeton, doing the collation, later put it plainly:
I had no idea what it meant or was used for.
Barbara BeetonShe asked. She consulted Charles Goldfarb, who created SGML. Nobody could tell her whether the entities came from one source or had simply been gathered from several.
-
14 March 2000
Unicode proposal N2191
“Proposal for Encoding Additional Mathematical Symbols.” The glyph is on the list, because it was on the STIX list, because it was on the ISO list, because it was in the Monotype catalog.
-
March 2002
Unicode 3.2 — U+237C
Encoded. From here it propagates automatically into every conformant system built afterward: operating systems, fonts, browsers, phones. A shape whose meaning was already unrecoverable becomes permanent infrastructure.
-
April 2022
Jonathan Chan reopens the file
A blog post traces the chain backward through the standards and finds the bottom of it missing. The glyph acquires a reputation as the most mysterious character in Unicode. Updates in June 2023 and August 2024 push the paper trail further back with catalog scans — deepening the record, never reaching an origin.
The azimuth thread
There is exactly one interpretation on record. Berthold's catalogs, and Holmqvist after them, gloss the glyph as azimuth or direction angle — a bearing, the angle of a thing from a reference direction.
It is plausible on its face. A right angle is a reference frame; an arrow could indicate a heading. Surveying, navigation, and astronomy all need such a mark.
But no one has produced a single published mathematical, scientific, or engineering text that uses it. Not one. The interpretation exists only in catalogs describing the sort — never in a document setting it.
So the azimuth reading is not a solution. It is one more undocumented assertion, older than the others, sitting at the end of the trail where an answer should be.
What is actually known
Most writing about this glyph blurs these three categories together. They are kept apart here on purpose.
Established
- It is encoded at U+237C, in Unicode since 3.2 (March 2002).
- It appears in Monotype's 1963 catalog as S9576, and in the 1972 mathematical catalog as S16139.
- It is absent from ISO 8879 (1986) and present in ISO/IEC TR 9573-13 (1991).
- It carries AFII glyph code D97C / 55676.
- Berthold and Holmqvist glossed it as azimuth / direction angle.
- Barbara Beeton, who put it into STIX, did not know what it was.
Claimed, unverified
- That it means azimuth or direction angle. Asserted by two typographic sources; never evidenced in use.
- That ISO drew it specifically from the Monotype set. Strongly suggested by the timing; not documented.
Unknown
- What it means.
- Who designed it, and for whom.
- What was ever typeset with it.
- Why Monotype cut a matrix for it.
- How it crossed from Monotype into ISO, and on whose initiative.
- Why it was renumbered in 1972.
Why keep the name
The interesting thing about ⍼ is not that it is obscure. Plenty of Unicode is obscure. It is that the system worked exactly as designed and still produced this: a mark propagated across six decades and four standards bodies by nothing but each participant's reasonable assumption that someone earlier had a reason.
Every layer faithfully preserved a meaning that was never there. That is worth naming something after.
Sources
- Jonathan Chan, “U+237C ⍼ RIGHT ANGLE WITH DOWNWARDS ZIGZAG ARROW” (April 2022, updated June 2023 and August 2024) — the primary investigation, with catalog scans.
- Wikipedia: Angzarr — consolidated timeline and the Berthold/Holmqvist references.
- codepoints.net: U+237C — encoding properties.
- ISO/IEC TR 9573-13:1991 — ISO AMS-A entity set, ed. Anders Berglund.
- Unicode proposal N2191, “Proposal for Encoding Additional Mathematical Symbols” (March 2000).
Found an error, or a document that pushes the trail back further? That is genuinely the point of this page. hello@angzarr.app