Synonym
*From Greek* sunōnumos, *'having the same name'*
'Synonym' is a word about words — a metalinguistic term that names one of the fundamental structural relationships in language. To have a synonym is to share a name, or at least to share what a name points toward. The word itself comes from Greek *sunōnumos*, compounding *sun-* (together, with) and *onuma* (name). It entered English in the fifteenth century via Latin *synonymum*, itself borrowed from the Greek. The journey is short and transparent. What is less transparent is the structural paradox this word carries inside it.
The -onym Family and PIE *h₁nómn̥*
The Greek *onuma* (also spelled *onoma*) belongs to one of the most durable reconstructions in comparative linguistics: PIE h₁nómn̥*, meaning 'name'. This root is not confined to one branch of the family. It runs through virtually every daughter language, arriving intact across millennia of phonological change.
Latin *nomen* (name, noun) gives English *nominal*, *nominate*, *nomenclature*, *pronoun*, and — through the shortening *noun* itself — the grammatical category that names things. Sanskrit *nāman* preserves the root in a form close to the reconstruction. Old English *nama* became Modern English *name*. Gothic *namo*, Greek *onoma*, Armenian *anun*, Old Irish *ainm*: the convergence is extraordinary.
The -onym Subfamily
Greek *onoma* became the productive suffix *-onym*, and from that suffix an entire metalinguistic vocabulary was assembled:
- antonym — the name that opposes - homonym — the same name for different things - pseudonym — a false name - anonymous — without a name - acronym — a name from initials - eponym — a name given to something from a person - patronym — a name derived from the father - toponym — a name derived from a place
Each term in this family names a different structural relationship between a sign and its referent, or between signs and other signs. The -onym words are the vocabulary of semiology — the discipline of signs — before semiology had that name.
The syn- Prefix and PIE *ḱom*
The prefix *syn-* (also *sym-*, *syl-*, *sys-* depending on the following consonant) derives from PIE ḱom*, meaning 'together, with, beside'. This root bifurcated in its descent: the Greek branch produced *sun-* / *syn-*, and the Latin branch produced *cum* / *com-* / *con-*.
The Greek line gives *sympathy*, *synthesis*, *synchronize*, *synagogue*, *symmetry* — all built on togetherness, combination, simultaneity. The Latin line gives *combine*, *commune*, *congress*, *connect*, *concur*, *contemporary*. English absorbed both streams, often holding parallel formations: *synthesis* and *composition* coexist; *synagogue* and *congregation* name the same kind of gathering through cognate prefixes.
'Same' itself belongs to this cluster, via Proto-Germanic sama*, which connects to the same PIE togetherness root. So *synonym* and *same* share an ancestor: to name a synonym is to identify something that is, in some sense, the 'same together'.
The Structural Irony of the Word Itself
Here the philological account meets its limit — and something more precise begins.
If the meaning of a sign is constituted not by any inherent property but by its differential relations within the system — if *big* means what it means because it is not *large*, not *enormous*, not *vast*, not *huge* — then two signs cannot occupy the same position in the system without one of them becoming redundant. The system does not tolerate perfect duplication. It converts any apparent equivalence into a distinction.
This is what the study of so-called synonyms invariably reveals. *Big* and *large* are frequently listed as synonyms. But speakers do not use them identically. 'A big deal' does not translate seamlessly to 'a large deal'. 'Large-hearted' differs from 'big-hearted' in register if not in reference. The moment you examine distribution carefully, the supposed equivalence dissolves into a network of finer contrasts: one term carries colloquial weight, the other a degree of formality; one combines freely with certain collocations, the other resists them.
Entire synonymy — two signs with fully identical value, interchangeable in every context — is structurally impossible. If value is relational, then two signs with the same value would be, in effect, one sign. The system would merge them or differentiate them. Languages do both: they lose one of a pair of true doublets over time, or they redistribute the pair across distinct semantic or stylistic territory.
The Paradox
'Synonym' therefore names a relationship that a rigorous structural account of language cannot fully endorse. The word exists. The concept is useful — useful enough to have persisted since antiquity. But the precision of the concept does not survive close examination. Every pair called synonymous, under scrutiny, turns out to be a pair of near-synonyms: overlapping but not coincident, equivalent in some contexts and differentiated in others.
This is not a failure of the concept. It is evidence that the system works exactly as structural analysis predicts. The -onym vocabulary, built on the most stable root in Indo-European, names the architecture of linguistic relations. And 'synonym' — the name for sameness — turns out, on inspection, to name a structural impossibility.
That is not nothing. That is how a language encodes its own theory of itself.