DANUBIUS · ISTER · ONE RIVER
Why does the Danube have two ancient names?
The Greeks first knew the lower course, from the sea — the Ister. From Central Europe came Danuvius. When the two maps joined, the ancients found they had named one river twice — yet the boundary between the names jumps hundreds of kilometres from author to author.
- Ister
- the Greek name of the lower course
- Danuvius
- the upper course's name, from the west
- no shared line
- each author draws the boundary elsewhere
Seven witnesses, two names
What each one calls it
Six centuries pass from Herodotus to Ptolemy; the boundary between the names moves from author to author — cataracts, Save, Illyricum, Axiopolis.
Herodotus, ~450 BC
“The Ister flows from the Celts through the middle of Europe and out to the Black Sea at Milesian Istria.”
Histories 2.33 (Godley transl.)
For Herodotus, Ister is the whole river — as he reconstructs it. He nails the mouth: five mouths, the greatest known river. He botches the source: Pyrene, among the Celts. What the Greeks knew from experience was the lower course.
Caesar, 52 BC
“…rectaque fluminis Danubi regione pertinet ad fines Dacorum — the Hercynian forest runs along the Danube down to the Daci.”
The Latin name shows up in Caesar, for the stretch from the west, down to the Daci. No boundary — and no Ister.
Strabo, ~AD 18
“Danuvius they called the upper part, from the sources to the cataracts… Ister the lower, from the cataracts to the Pontus.”
Geography 7.3.13 (Jones transl.)
One boundary at the cataracts: Danuvius above, Dacian country; Ister below, Getic country. One mental map among several — not the only ancient boundary.
Appian, on 35 BC
“…on the other side of the Ister, which is there called the Danube, but a little lower down is called the Ister. The Save flows into it.”
Illyrian Wars 22 (White transl.)
Inside the story of Octavian's campaign (35–33 BC): at Segestica and the Save it is still the Danube; the Ister begins “a little lower down”. The Romans learn the river as they go.
Pliny, ~AD 77
“…per innumeras lapsus gentes Danuvi nomine… et unde primum Illyricum adluit Hister appellatus…”
Danube “through nations innumerable”, Ister “where it first washes Illyricum”. The landmark is vague — Illyricum itself was an elastic notion. In Pliny the river takes sixty tributaries and leaves by six mouths.
Ptolemy, ~AD 150
“…that part of the Danube between the mouth of the Tibiscus and Axiopolis, from which town as far as the Pontus and the mouth of the Danube it is called the Ister…”
Geography 3.8.1, in Ionescu & Rovithis-Livaniou
In Ptolemy the Ister begins only at Axiopolis, by today's Cernavodă — hundreds of kilometres downstream of Strabo's line.
Tacitus, AD 98
“Danuvius… donec in Ponticum mare sex meatibus erumpat — out into the Pontic Sea by six mouths.”
Tacitus uses Danuvius for the whole river, mouths included: six mouths, plus a seventh through marshes. The upper name has swallowed the whole course.
How the maps joined
When two rivers became one
The Greeks knew the Ister from the sea; its sources stayed speculation. The Romans came from the west with Danuvius. In the generation of the Illyrian campaigns the two maps touched — and one river was left with two names.
The Greeks, from the sea
The Ister is known early at its mouths; the upper course stays speculation — Herodotus imagines the source among the Celts, at Pyrene.
The Romans, from the west
Caesar writes Danuvius (52 BC); Augustus pushes Illyricum “to the bank of the river Danuvius” (Res Gestae 30).
The joint
In the story of Octavian's campaign (35–33 BC), Appian already uses both names for one river: Danube at the Save, Ister “a little lower down”.
From the sources to the mouths
The course with two names
Three stops on one river: above, the boundary, below. The Roman map draws it as one line; the name cuts it — each author elsewhere.
The map that moves
Where did the Ister begin?
One river, six mental maps. Pick an author and watch the boundary jump hundreds of kilometres.
DANUVIUSISTER
STRABO (~AD 18): the boundary at the cataracts — Danuvius above, Ister below.
HERODOTUS (~450 BC): no boundary — the whole river is Ister, as he reconstructs it.STRABO (~AD 18): the boundary at the cataracts — Danuvius above, Ister below.APPIAN (on 35 BC): Danube at the Save, Ister “a little lower down”.PLINY (~AD 77): Ister “where it first washes Illyricum” — a vague zone, not a point (position approximate).PTOLEMY (~AD 150): Ister only from Axiopolis on.TACITUS (AD 98): no boundary — the whole river is Danuvius, mouths included.The river did not move its name. The authors' mental maps did not agree.
Vindobona is today's Vienna; the Save meets the Danube at Belgrade; the cataracts are the Iron Gates; Axiopolis is by Cernavodă. Positions approximate, proportional to the course.
| Author | Where the Ister begins |
|---|---|
| Herodotus | all Ister |
| Strabo | at the cataracts |
| Appian | just below the Save |
| Pliny | at Illyricum (vague) |
| Ptolemy | at Axiopolis |
| Tacitus | all Danuvius |
And Vienna? Old scholarship puts the division at Vindobona and attributes it to Agathemerus — but the claim could not be verified in the critical edition. On the map, Vienna stays what is certain: a landmark, not an attested boundary.
The sources
Tacitus starts it at Mount Abnoba; Șăineanu, in the Black Forest. Herodotus's Ister starts in the west too — but in imagined geography (Pyrene), not in mountains the Romans had surveyed.
The cataracts
Strabo moves the name at the cataracts — today's Iron Gates, below Belgrade. That is Strabo's boundary, not antiquity's: Appian, Pliny and Ptolemy draw the line elsewhere.
The mouths
Five mouths in Herodotus, six plus a seventh through marshes in Tacitus, three today (Șăineanu). The count belongs to the author, not the water.
The upper name
Danubius: the river that flows
Danubius comes from an old word for “river” itself: *dānu-. How sure each kin is, below — by steps, not by trust.
“Dérivation en –io- d'un th. i.e. *dānu- qui désigne le fleuve en avestique, dānu-, et en ossète don”
Delamarre 2012, in Boucherit 2014 RECONSTRUCTION
The confidence ladder
Avestan dānu-, Ossetic don ATTESTED
“River” — attested forms: in old texts and in today's speakers' mouths.
Danuvius and the Don RECONSTRUCTION
Our river and the Greeks' Tanais, from *dānu- “river”. The hard core of the family.
The Dnieper DISPUTED
Danapris to the ancients. Kretschmer proposes Scythian *Dānu-apara- “the river behind” — “tempting, but with phonetic difficulties” (Vasmer); Sobolevsky proposes *Danapras. Unsettled.
The Dniester DISPUTED
Danastris to the ancients. Kretschmer proposes *Dānu-nazdyō “the river in front” — Vasmer files the name under “unexplained”, with reservations (ě, r). Unsettled.
Eridanos DISPUTED
Delamarre reads it as Celtic *ēri-dānos, “the Western river”. A named proposal, not a consensus.
Which channel brought *dānu- to the Danube — Celtic (Zeuss 1837) or Iranian, through Scythians and Sarmatians — is still argued. The -uvius suffix stays open: no segmentation has authority.
Where we stand
What we know and what we reconstruct
Four drawers for the whole page: what is attested, what is reconstructed, why the reconstruction holds, and what stays unknown.
What we know
Ister is the Greek name of the lower course, Danuvius of the upper. Strabo cuts at the cataracts, Appian below the Save, Pliny at Illyricum, Ptolemy at Axiopolis. Histria is Milesian, 7th century BC.
What we reconstruct
The root *dānu- “river” and the hard core of its family. The Dnieper/Dniester segmentations and the transmission channel stay open.
Why it holds
The forms are alive today: Avestan dānu-, Ossetic don. But each kin has its own step of confidence — see the ladder above.
What stays unknown
The meaning of Ister: the Thracian label and the *is(t)r proposal, unsettled. And the road of Dunăre into Romanian: the Slavic or the native channel.
Three stops
Three dates
The whole way on one rail: the city, the witness, the boundary.
Histria
7th c. BC
Milesians found Istros on the bay by the Ister's mouths; Eusebius gives 657/656.
stop 1 of 3
Herodotus
~450 BC
One river, one name — in his reconstruction: Ister from the Celts to the Pontus, five mouths.
stop 2 of 3
Strabo
~AD 18
One boundary at the cataracts — Strabo's: Danuvius above, Ister below.
stop 3 of 3
Caesar 52 BC · Tacitus AD 98 · Ptolemy ~150 · Pârvan digs Histria from 1914.
The road of the names
Ister
the Greek name of the lower course
Danuvius · Ister
each author draws the boundary elsewhere
Dunăre · Istru · Histria
one living name, one bookish name, one place-name
The lower name, in brief
And what does Ister mean?
Three answers with names — and no settled meaning. The page's verdict: UNKNOWN
The Thracian
Zeuss (1837): Ister is the Thracian name of the lower course. A label, not a meaning — and that is still where things stand.
*is(t)r
Pârvan and Drăganu propose Thracian *is(t)r — “flowing, strong current” (via Eremia 2012). A named proposal, not a consensus.
Matoas
Stephanus of Byzantium knew one more name for the river: Matoas (in Smith 1854). An ancient curiosity, not an etymology.
No meaning of Ister is established: a gloss without a named linguist is invention.
And “the boundary at Vindobona”? Old scholarship attributes it to Agathemerus (Smith cites Agathemerus 2.4) — but the claim could not be verified in Müller's critical edition. For now: an attributed tradition, not verified testimony.
The Adriatic error: the idea that the Danube also reached the Adriatic spread in the 4th century BC, with Theopompus and the Argonauts; the Romans disproved it as they conquered the Danube lands, with fresh legion knowledge (Habaj 2018). On the Tabula Peutingeriana, the Save still flows into the Adriatic.
Zeuss 1837, in Ferguson 1862 →
What Romanian kept
Dunăre, Istru, Histria
One name lives on the river. The others stayed on the map and in books.
Dunăre: the Slavic channel
Șăineanu (1929): “[Derivat dintr'un primitiv; Dună (ung. DUNA, slav. DUNAVŬ) cu aceeaș finală analogică ca în (aiu)re]”. The handbook hypothesis — but not the only one.
Dunăre: the native hypothesis
Pârvan: Iranian dan- with Dacian -ris; Pușcariu: dan- with -re; Ivănescu: Thracian *Donaris, an action noun in -are (or pre-Indo-European dhen with -are). The unusual -re ending, unique among the neighbours, is what the dispute turns on.
Dunăre: what we know for sure
The Romanian form stands alone among its neighbours; the hydronym's family is ancient; the immediate road into Romanian stays disputed in the works checked. DE also records Donaris, “the Geto-Dacian name”.
dunăre, a common noun
DEX: “from the proper name Dunăre”. The river became a measure: dunăre de apă, dunăre de mânios, dunăre turbată. MDA2 gives the first attestation in Zanne (1895).
Istru, the bookish name
Șăineanu: “the Geto-Greek name of the Danube, especially of its lower course”. Today historical and bookish — the living river is called Dunăre.
Histria, the place-name
A Milesian city, 7th century BC; the first coins on Romania's territory were struck here (5th century BC); Pârvan digs from 1914. The ancient name Istros lives on as a place-name, not a river-name. Today's commune says Istria, the site Histria.
Every “dunăre” entry on dexonline →
The Dunăre dispute surveyed: Eremia 2012 (PDF) →
One river. Two old names. And no line antiquity ever managed to draw once and for all.
Test yourself
Where would you draw the line?
Three authors, three right answers. Place the boundary where you think each one puts it — the river is the same, the lines jump hundreds of kilometres.
Set the marker on the river's course, then press Check. Arrow keys ← → move the marker; fingers can drag it too.
Strabo ~AD 18
SourcesVindobonaSaveCataractsAxiopolisMouthsat the cataracts (the Iron Gates). Danuvius above, Ister below — Strabo's boundary, not antiquity's.
Appian on 35 BC
SourcesVindobonaSaveCataractsAxiopolisMouthsa little below the Save. Still Danube at Segestica; the Ister begins “a little lower down” — the Romans learn the river as they go.
Ptolemy ~AD 150
SourcesVindobonaSaveCataractsAxiopolisMouthsat Axiopolis (by Cernavodă). The Ister begins only toward the sea — hundreds of kilometres downstream of Strabo's line.
Score: 0 of 3
The right boundaries:
- Strabo — at the cataracts (the Iron Gates)
- Appian — a little below the Save
- Ptolemy — at Axiopolis (by Cernavodă)
As the authors drew them: one river, three lines.
Sources
The ancients first, then modern readings, the names and Romanian. Every risky claim points at its source.
The ancients
- Herodotus 2.33: Ister from the Celts to Istria (Perseus)
- Herodotus 4.47–48: five mouths, greatest river (Perseus)
- Caesar 6.25: fluminis Danubi (Latin Library)
- Strabo 7.3.13: the boundary at the cataracts (LacusCurtius)
- Appian, Illyr. 22: Danube at the Save, Ister below (Livius, White transl.)
- Pliny 4.79: Ister from Illyricum (Rackham transl., Attalus)
- Pliny 4.79: the Latin text (Wikisource)
- Ptolemy 3.8.1: Ister from Axiopolis (Hamilton/Falconer 1854 note)
- Tacitus, Germania 1: Danuvius (Latin Library)
- Res Gestae 30: Illyricum to the Danube (Latin)
Modern readings
- Herodotus's Istros: Dan 2011, Dacia 55 (HAL, PDF)
- The two-mouthed Danube: Habaj 2018 (DOI)
- Appian and Dio on the Illyrian wars: Šašel Kos (PDF)
- Histria, marine bay and lagoons: Hanganu 2012 (summary)
The names
- Don: Vasmer
- Dnieper: Vasmer
- Dniester: Vasmer
- Delamarre in Boucherit 2014: *dānu- (PDF)
- The Dunăre dispute: Eremia 2012 (PDF)
- Istros/Histria: Princeton Encyclopedia (Perseus)
- Axiopolis: Princeton Encyclopedia (Perseus)
Romanian
- dunăre/Dunăre: DEX, MDA2, Șăineanu, DE, on dexonline
- Istru: Șăineanu, on dexonline
- Histria: DE, on dexonline
- Neacșu's letter: Dima 2021 (study)
- Neacșu's letter: the English translation (Wikisource)
- Neacșu's letter: the transcription (Wikipedia)
Scholarship history
- Smith 1854: Danubius, late usage (Perseus)
- Smith 1854: Vindobona (Perseus)
- Ferguson 1862 after Zeuss 1837: Celtic/Thracian (Gutenberg)
Neighbours
Type: Fraunces and Source Serif 4, under the SIL Open Font Licence.