Manuel de Codage Explained: How Egyptologists Encode Hieroglyphic Text and Layout
Before Egyptian hieroglyphs could be typed directly as Unicode characters, Egyptologists still needed a way to enter them into computers.
A hieroglyphic text presented two problems at once.
First, each sign had to be identified.
Second, the computer had to know how those signs were arranged:
side by side
stacked vertically
grouped into a quadrat
enclosed in a cartouche
broken across lines
The Manuel de Codage, usually abbreviated MdC, was created to solve that problem with ordinary keyboard characters.
Instead of storing a picture of a seated man, a water ripple, or an owl, an Egyptologist could type sign identifiers and punctuation-like operators.
Software could then interpret the code and draw the hieroglyphic text.
That made MdC enormously important for digital Egyptology.
But it is easy to misunderstand what the system is.
MdC is not the ancient Egyptian language.
It is not a new transliteration alphabet.
It is not the same thing as Unicode.
It is a computer-readable notation for identifying hieroglyphic signs and describing how they should be arranged.
Why Hieroglyphs Needed More Than a Simple Character List
Alphabetic text is mostly linear.
You type:
A B C D
and the letters appear one after another.
Hieroglyphic writing is more spatial.
Signs can be:
- placed beside one another;
- stacked above and below one another;
- fitted into compact rectangular groups;
- inserted into larger signs;
- written inside cartouches and other enclosures;
- oriented according to text direction.
The visual unit traditionally called a hieroglyphic quadrat can contain several signs occupying different parts of the same block.
A computer therefore needs more than:
sign 1, sign 2, sign 3
It needs something closer to:
place sign 1 above sign 2, then put that group beside sign 3.
MdC supplied a plain-text grammar for expressing those relationships.
MdC Turned Hieroglyphs Into Plain-Text Instructions
The system grew out of the computerization of Egyptology in the 1980s.
A committee involving Jan Buurman, Nicolas Grimal, Michael Hainsworth, Jochen Hallof, and Dirk van der Plas developed a common approach for computer input of hieroglyphic texts. The published Manuel de Codage became highly influential in Egyptological software.
Its central idea was practical:
identify signs with ordinary alphanumeric codes, then describe their layout with ASCII punctuation.
The published 1988 manual suited an era when plain ASCII was vastly easier to exchange between computers than specialist hieroglyphic graphics. A code such as A1:N35 could travel as ordinary text even when the receiving computer did not yet have a sophisticated Egyptian font. That portability was one of MdC’s great practical strengths.
One major source of sign identifiers was Gardiner’s Sign List.
Codes such as:
A1
D36
G17
N35
identify entries in Gardiner’s catalog.
The letter names a broad sign category; the number selects a sign within it.
As explained in Gardiner’s Sign List, these are catalog identifiers, not pronunciations.
So:
A1 = which sign
not:
A1 = how the sign was read aloud.
This separation between sign identity and linguistic value is fundamental to MdC.
Some Signs Can Be Entered by Phonetic Code
MdC also allows many common phonetic signs to be entered through ASCII-friendly transliteration codes.
That can be faster than remembering the Gardiner number for every sign.
A sign might therefore be represented either by:
its catalog identifier
or:
an MdC phonetic shortcut
when the system defines one.
For example, the sign identified as A12 can also be represented in MdC by the phonetic code mSa, corresponding to an Egyptological reading convention.
That convenience has a limit.
One transliteration value can be associated with more than one hieroglyphic sign. A shorthand code therefore cannot always express every graphic distinction that a precise sign identifier can.
For exact sign-level work, a Gardiner-style identifier is often less ambiguous because it tells the software which cataloged sign is intended.
This does not mean every transliterated word can simply be typed and automatically converted into the correct hieroglyphic spelling.
MdC phonetic codes identify particular signs.
Egyptian words still require decisions about:
- which signs are actually written;
- phonetic complements;
- classifiers;
- spelling variants;
- layout.
So:
MdC phonetic code ≠ automatic Egyptian word translation.
Three Core Operators Describe Basic Quadrat Layout
MdC uses simple punctuation to describe sign relationships.
- separates successive quadrats
An expression such as:
A1-B1
places the two signs in successive groups.
The hyphen is not Egyptian punctuation.
It is a computer instruction separating quadrats.
* joins signs horizontally inside a quadrat
A1*B1
places the two signs beside one another inside one block.
This is not the same as A1-B1, where they occupy successive groups.
: stacks signs vertically
A1:B1
places one sign above the other inside a shared quadrat.
The distinction between * and : is one of MdC’s most important ideas because it preserves two-dimensional structure rather than reducing the text to a flat character sequence.
Parentheses Make Complex Groups Unambiguous
Parentheses control grouping and precedence.
For example:
(A1*B1):N35
first treats A1*B1 as a horizontal unit, then stacks that unit relative to N35.
This works like grouping in mathematics or programming.
The same signs can produce different structures if the parentheses move.
A real MdC sequence such as:
I10:D46-M17-N35
can therefore be read in stages:
I10:D46 — one vertically stacked quadrat
– — move to the next quadrat
M17 — next sign group
-N35 — another successive group
The exact glyph sizes and spacing are chosen by the renderer.
MdC preserves the structural relationships needed to reconstruct the grouping.
MdC Encodes a Logical Layout, Not a Photograph
This distinction is essential for archival work.
An MdC expression can tell a renderer:
these two signs form one horizontal group
this group sits above another sign
these quadrats follow one another
But the code does not automatically preserve every visible property of the original monument.
The source may contain:
- unusually stretched signs;
- local artistic variants;
- nonstandard spacing;
- damage;
- carving depth;
- paint;
- slight rotations;
- hand-adjusted proportions.
Some later MdC dialects added controls for part of this detail, but the basic notation is primarily structural.
That makes MdC useful for typesetting and text processing.
It also means a photograph, facsimile drawing, and MdC transcription are not interchangeable records.
The photograph preserves the object.
The MdC file preserves an interpreted sign sequence and layout model.
A researcher checking a difficult reading should therefore return to the original image or publication rather than assuming the encoded file contains every graphic nuance.
Later MdC Practice Became More Complicated Than the Core Syntax
The MdC tradition grew beyond -, *, :, and parentheses.
Different editors added ways to represent features such as:
- cartouches and other enclosures;
- line and page endings;
- red and black text;
- shading;
- rotations;
- mirrored signs;
- ligatures;
- blank space;
- damaged or philologically marked text.
This created a portability problem.
Serge Rosmorduc’s documentation of later MdC practice notes that programs often extended or modified the original standard. Features associated with WinGlyph, Tksesh, MacScribe, and other tools were not always identical.
So a historical file described as “MdC” may actually contain a software dialect.
A basic sequence using Gardiner codes and the common layout operators is broadly intelligible.
A highly elaborate file may require knowledge of the program that created it.
This also means MdC should not be mistaken for an exact photographic description of a monument.
It can encode relationships such as:
above
below
beside
grouped together
but a renderer still chooses exact size, spacing, centering, and proportions.
Two programs may therefore draw the same MdC expression somewhat differently while preserving the same underlying structure.
MdC Is Not Transliteration
Egyptological transliteration represents the linguistic reading of Egyptian writing with Latin-based scholarly characters.
For example, a word might be transliterated:
nfr
MdC can use ASCII letter combinations that resemble transliteration, and many phonetic sign shortcuts derive from transliteration values.
But the two layers should not be confused.
Transliteration answers:
How do Egyptologists represent the linguistic value of this written Egyptian?
MdC answers:
Which hieroglyphic signs and layout relationships should the computer reproduce?
The distinction is explained more broadly in Egyptian hieroglyph transliteration.
One Egyptian word can have multiple historical spellings while sharing the same normalized transliteration.
MdC needs to preserve the actual written sign choices if the goal is faithful sign-level encoding.
MdC Is Not Unicode
This is the most important modern distinction.
In MdC:
A1
is ordinary Latin text used as a sign identifier.
A specialized program interprets that code and draws the corresponding hieroglyph.
In Unicode, the hieroglyph itself can be stored as a standardized encoded character.
The layers are therefore different:
MdC → markup interpreted by Egyptological software
Unicode → standardized character code points and format controls
A Gardiner identifier such as A1 is not a Unicode code point.
It is a catalog label.
Likewise, the visible drawing is supplied by a font or renderer rather than stored inside the MdC code itself.
That means three things remain distinct:
sign identity
encoded character
visible glyph
Different fonts can draw the same identified sign differently.
Unicode preserved several MdC layout ideas
Unicode did not eliminate the two-dimensional layout problem.
The Unicode Standard’s current Egyptian hieroglyph specification explicitly preserves several concepts familiar from MdC in its Egyptian Hieroglyph Format Controls:
MdC : → vertical joining
MdC * → horizontal joining
MdC parentheses → segment grouping controls
Modern Unicode also supports layout operations beyond the simplest MdC model, including insertion and overlay.
The relationship is therefore evolutionary.
MdC established a practical plain-text way to describe hieroglyphic grouping; Unicode later standardized Egyptian characters and layout controls at the character-encoding level.
The current workflow is discussed in typing and copying Egyptian hieroglyphs with Unicode.
A Correct MdC String Can Still Encode a Wrong Reading
MdC records a digital representation.
It does not prove the Egyptological interpretation is correct.
A file can be syntactically valid while containing:
- a misidentified sign;
- the wrong variant;
- an incorrect word division;
- a mistaken restoration;
- a poor transliteration;
- a bad translation.
So MdC should be checked against the source image or publication when accuracy matters.
The evidence-chain method in checking a hieroglyphic translation still applies.
This matters especially for older digital corpora.
Many projects were created before modern Unicode hieroglyphic support, so MdC remains common in research archives, teaching files, legacy editors, databases, and conversion pipelines.
Its plain-text appearance does not guarantee easy conversion.
A converter may need to understand sign identifiers, grouping operators, extensions, and the exact MdC dialect used by the original software.
A Safe Way to Read an MdC String
When you encounter MdC, analyze it in layers.
1. Identify the sign codes
Separate Gardiner codes and recognized phonetic shortcuts from operators.
2. Find the quadrat boundaries
Hyphens normally separate successive groups.
3. Resolve internal layout
Asterisks indicate horizontal grouping and colons indicate vertical stacking.
4. Apply parentheses
Determine which subgroups belong together before rendering the larger structure.
5. Identify extensions
Cartouches, shading, rotation, red text, damage, or other markup may depend on the software dialect.
6. Render before interpreting linguistically
First confirm that the MdC reproduces the intended signs and layout.
Then transliterate and translate.
This prevents the digital syntax from being mistaken for the Egyptian language itself.
MdC, Transliteration, and Unicode Solve Different Problems
| System | Main job | Example |
|---|---|---|
| Gardiner code | Identify a cataloged sign | A1, N35 |
| MdC | Encode signs plus structural layout | A1:N35-B1 |
| Transliteration | Represent linguistic reading | nfr, pr, nswt |
| Unicode | Encode standardized characters and format controls | Actual hieroglyph characters |
| Font / renderer | Draw the visible glyphs | Screen or print shapes |
These systems interact, but they should not be collapsed.
A Unicode character can have a Gardiner-style identity.
An MdC code can use a phonetic shortcut.
A renderer can convert MdC into visible signs.
A linguistic tool can derive transliteration from an encoded sign sequence.
Each layer answers a different question.
Manuel de Codage Made Hieroglyphic Computing Practical
MdC belongs to an earlier stage of digital text technology, but its core insight remains important.
Hieroglyphic text cannot be represented adequately as a flat list of pictures.
A useful digital system must preserve both:
sign identity
and:
spatial relationships between signs.
MdC achieved that with compact ASCII notation: Gardiner codes identified signs, while hyphens, asterisks, colons, and parentheses described their structure.
Later software extended the system, and Unicode now provides standardized Egyptian characters plus dedicated format controls. Several of those controls preserve concepts inherited directly from MdC.
For the HieroglyphicsHub translator, the lesson is practical.
The same Egyptian text can exist in several digital layers:
source image
→ identified signs
→ MdC or another structural encoding
→ transliteration
→ translation
→ Unicode rendering
Confusing those layers creates errors.
Keeping them separate makes digital hieroglyphic text far easier to verify, convert, preserve, and read.
