Skip to content

Heterogram (Linguistics)

Use a written form inherited from one language as a conventional sign for a different-language word or phrase that readers supply aloud.

Version
v2 · 2026-10-03 · History
Domain-specific #
13301
Domain group
Humanities
Origin domain
Linguistics & Semiotics
Subdomains
Alloglottography, Ancient Writing Systems → Linguistics & Semiotics
Aliases
Linguistic heterogram

Core Idea

A linguistic heterogram is a written form inherited from one language's graphic tradition that a text in another language uses as a conventional sign for a local-language reading. The crucial relation is written donor form, spoken matrix-language equivalent. It is not enough that two languages share a script or that a word is visibly borrowed: the foreign-derived spelling must function in the matrix text with a different lexical reading. This is why the same item can look Aramaic on a Pahlavi page yet be read as Middle Persian.[1][2]

Book Pahlavi supplies a documented case. Aramaic written forms appear alongside more phonetic Middle Persian spellings, and the Frahang ī Pahlawīg records many such forms with Persian readings. A conventional reader need not be fluent in Aramaic merely to use the heterographic convention; the visible origin of the form is not evidence that an Aramaic word was spoken in that passage.[1][3]

Structural Signature

  1. Matrix-language text: the surrounding utterance or passage has a language in which it is understood and, when read aloud, realized.
  2. Foreign-derived graphic form: a word, sign or group of signs comes from another linguistic writing tradition. The donor form's historical origin is relevant even if current users no longer know the donor language.[1]
  3. Local lexical target: the reader supplies a native-language word or phrase with the intended meaning or grammatical role.
  4. Conventional mapping: scribes and readers share an orthographic rule connecting that form to its local reading; the form is not merely a one-off undeciphered string.[3]
  5. Contextual reading: grammar and writing convention determine how the item functions in the actual matrix-language sentence. Donor spelling need not directly encode the local pronunciation.[2]

Condensed: donor-derived written form + matrix-language text + conventional local reading = linguistic heterogram.

Sig role-phrases: matrix-language passage; inherited donor graphic form; conventional local lexical reading; evidence of the mapping.

What It Is Not

  • Not a spoken loanword by default. A borrowed word enters a language's lexicon and can be pronounced as part of that language; a heterogram may remain purely graphic while readers supply a native equivalent.[1]
  • Not a code-switched donor-language utterance. If the passage actually changes languages and pronounces the foreign word as such, the defining written/read mismatch is not established.
  • Not every logogram. A monolingual writing system can use signs for words; the heterogram additionally has a donor-language graphic origin and a different-language conventional reading.
  • Not every morphogram. A morphogram represents a morpheme through a graphic meaning relation; heterograms may stand for a word or phrase, and the cross-language provenance is constitutive here.
  • Not proof that the scribe was bilingual. Specialist work on Pahlavi says scribes could learn the system without knowing Aramaic.[1]
  • Not the literary “heterogram.” That unrelated term can mean a string with no repeated letters. This entry is about cross-language writing and reading.
  • Not necessarily an archaism in every period. A historical tradition may preserve an old written form, but age is a consequence of some practices, not the identity criterion.

Scope of Application

In Middle Iranian philology, Pahlavi texts contain Aramaic heterograms, also termed Aramaeograms or, in related literature, ideograms. The written Aramaic preposition \(ʿL\), for example, was read as Middle Persian \(ō\). That pairing makes the cross-language mapping visible: donor-language graphic material occupies the page, but the locally supplied word belongs to the matrix language.[4][1]

The Frahang ī Pahlawīg lists mostly Aramaic written forms with phonetic spellings of corresponding Persian words. Its verb list, for example, pairs \(ʿZLWNtnˈ\) with \(šwtnˈ\), interpreted as šudan, “to go.” The mapping is visible rather than merely asserted. Yet MacKenzie cautions that a glossary lemma alone does not prove that the heterogram occurs in a surviving running text; some listed forms are unattested or disputed. The number and distribution of attested forms are historical facts about that tradition, not required counts for any other heterogram system.[3]

In Hittite writing, original linguistic research compares Sumerograms and Akkadograms in texts whose spoken matrix language was Hittite and proposes heterogram as a useful cover term. The core relation transfers, but individual signs and readings should be checked in their specific source before claiming that every occurrence works identically.[5]

Modern written Latin abbreviations such as “i.e.” sometimes receive an English expansion when read aloud, and “&” can stand for a local word meaning “and.” Those are plausible analogues, but the classical philological term is anchored most securely in the ancient writing traditions above. This entry does not depend on treating every modern abbreviation as an admitted heterogram.

Clarity

The notation on a page and the voice in which a passage is read are different objects. A historian who transliterates the graphic sign as its donor-language form has recorded the written layer; a historian who reconstructs the local-language spoken equivalent has recorded the reading layer. Confusing them can create a phantom loanword. Conversely, a reader who simply does not know the local value cannot infer it from graphic appearance alone; a convention, bilingual gloss or contextual philological evidence is needed.[1][3]

The Pahlavi case also corrects a tempting claim in the seed: “readers must know both the donor spelling and the native word.” They need know which graphic pattern conventionally maps to which local value, but not necessarily the donor language's vocabulary or pronunciation. Specialist accounts explicitly caution against inferring Aramaic proficiency from heterograms in Middle Persian texts.[1]

Manages Complexity

The distinction separates script history, orthographic convention and spoken lexicon. Without it, a text with many Aramaic-looking signs might be misclassified as a text containing many spoken Aramaic words. With it, the analyst can preserve the donor graphic provenance while correctly reconstructing a Middle Persian or other matrix-language utterance. A glossary becomes a mapping tool rather than a list of direct spoken borrowings.[3][2]

The cost is philological dependence on transmitted reading knowledge. A sign can be historically obvious as foreign yet locally ambiguous in value; later glossaries or parallel spellings may resolve some forms, not all. Labeling a sign “heterogram” identifies the mapping type but does not automatically decode the exact lexical target in every manuscript.

Abstract Reasoning

Begin with a specific inscription or passage and state its matrix language. Identify the disputed graphic form and evidence for its donor-language origin. Then find evidence for how readers in that textual tradition supplied a local-language equivalent: glossary, phonetic alternative, grammar or attested comparison. Test the counterfactual: if the graphic form were pronounced as the donor-language word or served as a native orthographic spelling, would the claimed heterogram identity still hold? It would not. Record uncertainty separately when the local reading is reconstructed rather than directly attested.[3][5]

This method supports a cleaner question than “is the text bilingual?”: What language is written in the sign's inherited shape, and what language is represented in the passage's reading? The two can differ without the scribe being a speaker of both.

Knowledge Transfer

Pahlavi Aramaeograms and Hittite Sumerograms show the same structural split across different scripts and historical settings: a donor-derived written sign is read as a local lexical item. The broad insight transfers to comparison of writing systems, but not every claim about a given Aramaic sign transfers to a Sumerian one. Orthographic persistence, scribal training and exact pronunciation have to be established within each tradition.[1][5]

The live Sign entry is the strict genus: a heterogram is one written form conventionally standing for a different local-language reading. Writing System denotes a whole inventory and rules, not this individual sign; Morphogram imposes a morpheme-level condition that need not fit every case. This is the linguistic heterogram, not the same-name literary word game.[1]

Examples

Aramaic ʿL in Middle Persian

An Aramaic-origin written \(ʿL\) occurs in Pahlavi and is read \(ō\) in Middle Persian. The local spoken value, not the shape of the sign, supplies its role in the Iranian utterance.[4]

Mapped back: matrix language = Middle Persian; donor graphic form = Aramaic \(ʿL\); conventional reading = \(ō\); function = an orthographic sign for the local word.

Pahlavi glossary pair

The Frahang ī Pahlawīg writes \(ʿZLWNtnˈ\) opposite \(šwtnˈ\), the Persian šudan, “to go.” The source also lists finite variants, such as \(ʿZLWNytˈ\) opposite \(šwytˈ\). This is a concrete written/read contrast in an attested glossary, not a fabricated sentence containing the form. Its occurrence in surviving running prose requires separate evidence.[3]

Mapped back: matrix language = Book Pahlavi/Middle Persian; donor form = \(ʿZLWNtnˈ\); local target = \(šwtnˈ\), šudan; mapping evidence = paired glossary lemma rather than a passage attestation.

Spoken borrowing as a near miss

Suppose a foreign word enters ordinary speech and is written in a familiar adapted spelling. Its historical source is foreign, but a reader pronounces the borrowed lexical item itself, rather than substituting a different local word for a retained donor-written form. Foreign etymology alone therefore does not make it a heterogram.

Structural Tensions

Written continuity versus phonetic transparency. Retaining inherited signs preserves the conventional graphic repertoire, but the local sound must be learned or glossed. Replacing them with phonetic Persian spelling makes the reading more immediately recoverable but removes the heterographic layer used across that scribal tradition. Diagnostic: for this passage and audience, is continuity with that written repertoire or direct recovery of the Persian reading more important? The Frahang's side-by-side pair makes the costs concrete.[3]

Structural–Framed Character

The spectrum runs from a readily recognized foreign spelling read aloud as itself, through a borrowed lexical item, to a donor-derived sign for a wholly local reading. The heterogram sits at the last end; visual foreignness alone cannot place an item there. Its status is descriptive rather than evaluative: it does not make a text more or less authentic, and it does not imply that a scribe knew the donor language. Human reading practice matters because a convention must connect the graphic form to the local word; the same marks quoted in an Aramaic passage would not be the Middle Persian heterogram. The origin of this relation lies in particular scribal and manuscript traditions, not in a universal script feature. The vocabulary heterogram travels across Pahlavi, Sogdian and cuneiform studies only when the written/read split is independently established; importing a Pahlavi sign's actual reading into Hittite is not recognition of a shared structure. Its character: a historically situated cross-language orthographic relation, identified by a donor-derived form's conventional matrix-language reading, not by graphic shape, prestige or bilingualism alone.[1][5]

Structural Core vs. Domain Accent

The portable skeleton is a surface inherited from one code that is interpreted under another; the relation is not the same as ordinary direct reading of visible signs. The live Sign genus captures the conventional visible-form-to-meaning relation, while the domain-bound mechanism requires an identifiable donor-language graphic tradition, a matrix-language passage and a conventionally supplied local lexical reading. Symbolic Representation or Translation may illuminate aspects, but neither is the selected parent; calling the heterogram itself a prime would overgeneralize that philological mechanism. A future prime question would ask whether cross-code reading of inherited signs recurs outside writing without erasing these evidence requirements.

This entry is a kind of Sign.

  • Symbolic Representation: a conventional visible mark stands for linguistic content.
  • Translation: the reader's local equivalent resembles a lexical translation, though the operation is an orthographic reading convention, not translation of a whole passage.
  • Context: text language and scribal tradition determine the sign's interpretation.

The live Sign entry is the strict genus; Symbolic Representation, Translation and Context remain related but are not separately asserted as parents.

Relationships to Other Abstractions

Local relationship map for Heterogram (Linguistics)Parents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.Heterogram(Linguistics)DOMAINDomain-specific abstraction: Sign — is a kind ofSignDOMAIN

Current abstraction Heterogram (Linguistics) Domain-specific

Parents (1) — more general patterns this builds on

  • Heterogram (Linguistics) is a kind of Sign Domain-specific

    A linguistic heterogram is a conventional written sign for a different local-language reading.

Hierarchy path (1) — routes to 1 parentless root

Neighborhood in Abstraction Space

Heterogram (Linguistics) sits in a sparse region of the domain-specific corpus (97th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (2551 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-08

Not to Be Confused With

A loanword is a spoken or lexical borrowing; a code-switch changes the language used in an utterance; a logogram need not have a foreign-language source; a morphogram is organized around a morpheme relation; a literary heterogram concerns repeated letters. The linguistic heterogram here specifically combines donor-derived written form with a different-language conventional reading.[1][5]

References

[1] Encyclopaedia Iranica, “Huzwāreš”. Specialist discussion of Aramaic heterograms in Pahlavi and scribal-language inference. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l

[2] Encyclopaedia Iranica, “Aramaic i. General”. Heterograms as Aramaic written words understood as Iranian equivalents. registry ↩a ↩b ↩c

[3] Encyclopaedia Iranica, “Frahang ī Pahlawīg”. Glossary of Aramaic heterograms with phonetic Persian readings. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h

[4] Encyclopaedia Iranica, “Avestan Language i. The Avestan Script”. Example \(ʿL\) read as Middle Persian \(ō\). registry ↩a ↩b

[5] Original research on Sumerograms and Akkadograms in Hittite texts. Publisher record and abstract used for scope; individual sign readings not inferred. registry ↩a ↩b ↩c ↩d ↩e