A household word with a prehistoric stem
Water is the ordinary English name for the liquid that falls as rain, fills rivers and seas, and, when the temperature shifts, becomes ice or steam. The same noun covers a glassful, a stretch of open sea, and, in older medical English, several bodily fluids. Speakers were already using the headword in Old English, and the core sense has barely moved.
The origin of the word water is native inheritance, not a loan from Latin or French. English carried the form up from West Germanic.
Old English and the West Germanic line
In Old English the noun was wæter, usually declined as a strong neuter, though a few masculine, feminine, and weak-looking forms turn up. The Oxford English Dictionary places the earliest English use in the Old English period, before 1150. A related verb, wæterian, is just as old: to moisten, to irrigate, to supply drink, or to lead cattle to a stream. Middle English kept the stem, with spellings such as water, watere, and watir, until the modern spelling settled.
This is a West Germanic word. Dutch still has water. German has Wasser, from Old High German wazzar. Old Saxon had watar; Old Frisian had wetir. Those forms point back to Proto-West Germanic watar and, behind that, Proto-Germanic watōr. Handbooks do not all write the Germanic ancestor the same way: some use a fuller watōr, others a shorter watr-type reconstruction. The disagreement is about how to notate the stem, not about whether English belongs in this family.
Gothic watō, recorded in the east of the Germanic world, is a close cousin that had begun to follow an n-stem pattern. North Germanic took a sharper turn. Old Norse vatn, Icelandic vatn, and Swedish vatten keep an n that West Germanic largely lost. That split is not a later borrowing between neighbors. It is a fossil of an older Indo-European habit in which one noun used an r-stem in some cases and an n-stem in others.
The r/n split
Reconstructed Proto-Indo-European wódr̥ named the substance itself. It belongs to a small class of heteroclitic neuters: the nominative and accusative ended in r, while other cases used n. Hittite still shows a split of that kind, with wātar beside n-forms elsewhere in the paradigm. Classical Greek kept a similar pair in húdōr and its oblique stem húdat-. Germanic rebuilt the whole pattern. West Germanic leveled the r through the word, which is why English ends in r. North Germanic mostly leveled the n, which is why the Scandinavian forms look like vatn. A rare Old Icelandic vatr shows that the r-form was not unknown in the north, only that it did not win.
The root underneath is wed-, covering both the liquid and the quality of being wet. English wet is not derived from water; it is a sibling from another vowel grade of the same root. Otter comes from a related formation that once meant an animal of the water. The Greek descendant later supplied English with the combining form hydro- in hydrant, hydrogen, and hydraulics, so the learned prefix and the native noun are distant relatives, not a pair coined in a chemistry lab.
Some linguists reconstruct two old Indo-European words for the liquid: wed- for the inanimate stuff and a separate root, often cited as ap- or h₂ep-, for water treated as a living force. They draw a parallel with two roots for fire. That contrast is a scholarly inference about prehistoric vocabulary, not a fact of historical English. A further comparison with Uralic words such as Finnish vesi is also disputed: loan, chance likeness, or deeper link remain open.
Relatives from Hittite to whiskey
Once vowel grades are allowed for, the family is wide. Lithuanian vanduo and Slavic voda sit on the same base; vodka is a diminutive of that Slavic noun. Latin unda, used chiefly of a wave, is usually tied to the same root, though the path is less tidy than the Greek match and is sometimes given with a hedge. Early Irish uisce, the source of whiskey through usquebaugh, comes from an extended form of the old oblique stem. Sanskrit udan belongs to the zero-grade side of the same noun. Hittite wātar is often cited because, after millennia, it still looks startlingly close to the English word.
None of these languages borrowed the English form. They inherited parallel descendants of one prehistoric word. Inside modern English the morphology is simple: water is not a prefix plus a root plus a suffix. The interesting structure sits one stage earlier, an r/n stem built on wed-.
Later senses without a new root
The chemical formula is modern; the everyday meaning is not. Classical and alchemical English could treat water as one of the four elements, beside earth, air, and fire, without changing the word. By the 1650s, waters in the plural often meant the seas of a region, and especially a state's maritime claims. Head above water begins as a drowning image in the seventeenth century and is recorded figuratively, for being out of difficulty, in the eighteenth.
The verb widened in other directions. Late Middle English already used it for dilution; water down, in the sense of weakening a statement or a product, is nineteenth-century. A mouth that waters, as a sign of appetite, appears in the fifteenth century, as does make water for urination.
A separate noun sense, the water of a diamond, meaning its quality or luster, is attested around 1600. One suggestion, offered cautiously in older dictionary commentary, is that it calques Arabic māʾ, which can mean both the liquid and brilliance. That remains a possibility, not a closed derivation. British English also keeps a countable use in which a water is a lake or pond, a survival of the older body-of-water sense that river names still show.
What Icelandic still shows
If the headword feels too simple to have a history, the Scandinavian cognates are a useful check. English kept the r of the old nominative. Icelandic vatn kept the n of the old oblique cases. Same noun, split at the Germanic stage, still visible in two living vocabularies. That split, not a borrowed classical lookalike, is the most concrete fossil the English word still carries.