Oxford University Press's
Academic Insights for the Thinking World

Muddy waters

When a word is isolated, etymologists are in trouble. A typical example is Engl. hunt, discussed last week (the post for February 12, 2020). Its English neighbors (hent, hint, and hand) shed little light on how hunt originated. No Germanic cognates exist (which is odd); all the other languages have different, partly obscure verbs for hunting. A non-Germanic Indo-European related verb, if it exists, should begin with k, and Greek kíein “to go, move,” which, naturally, occurred to our earliest etymologists, especially because it is the root of the Greek verb meaning “to hunt,” has a root vowel incompatible with short u in Old English. It is part of its own family: Latin citus “rapid, swift,” and many others (see it in cite, cinema, kinetics, etc.: cinema is really kinema). But often, the cognates are so numerous that researchers are lost, embarrassed by the riches they face. This is what happens when we begin to investigate the origin of the English word mud.

Its immediate neighbors are plentiful. The first occurrences of mud in texts go back to the fourteenth century, and that is why it is believed to be a borrowing from Middle Dutch or Middle Low German, even though a word of such semantics might easily exist for centuries, without making its way into books: one needs a special context for it. Dutch has modder, corresponding to German Moder and northern German Modder; in their vicinity, we find old and late words like mode, mudde, mod, modd, and many others.

Moder (a German word mentioned above), like Engl. mud, seems to have the root mo– or mu– and some suffix (or extension, as consonants of this type are called in etymological dictionaries). Once we are left with such a short root, we encounter an overwhelming number of seemingly related formations. If we begin with the Scandinavian mo– words (including those occurring in dialects), we will end up with “sand, gravel; grain; dust, haze, cloud; clay, peat,” and a few other nouns whose sense has strayed too far from any one listed here. A great multitude of such words was the subject of an investigation by the Swedish linguist Ivar Lundahl (1931). He suggested that the original sense of the mo– words was something like “crushed, broken matter.” I might add “loose earth” to it. Mud and its look-alikes will easily fit this definition.

In the morass of despair. Despair by Edvard Munch. Public domain via Wikimedia Commons.

Next to such mo-words, we find Engl. moor “a tract of unenclosed waste ground,” morass “a wet swampy tract,” and marsh “low-lying watery land.” Each has close relatives outside English, and their history is known rather well. Morass, for example, traveled from Germanic to French and back to English. All of them look like belonging together and being related to Swedish mo “a tract of sandy ground, with some pine trees on it.” (Moberg is a well-known family name; berg “mountain”). The oldest Germanic root could not be mo– or mu-, so that the full form is usually reconstructed as mu-ha or mu-wa. Not only morass and marsh have something to do with water (the list in Lundahl’s paper is very long), and it has been suggested that the initial meaning of the root was “wet ground; wetness.” This formulation can be found in our best etymological dictionaries, but we also face many mo-words, referring to small fragments (grain, sand), surfaces in which such fragments are especially conspicuous, and substances whose structure is not solid (like mud). They are especially common in the dialects of northern Germany, Dutch, and Scandinavian. Since in English, words like moor seldom occurred in the oldest period, the borrowing of mud remains a possibility.

A good, old-fashioned marsh. Brandon Marsh by Black Kite. CC-SA-3.0, via Wikimedia Commons.

Lundahl made a strong effort to disprove the idea that the initial root of the words being discussed here was “wetness,”and indeed “sand, dust, grain” do not fit this idea, even if “haze, cloud, marsh,” and a few others do. There may be an additional argument against “wetness, wet ground” being the predominant sense of the mo– root. In the Scandinavian myth of the end of the world, the forces of destruction (giants) clash with the gods, and all the contestants perish. The leader of the giants is called Muspell. This partly enigmatic name was known far and wide, because it also occurs in the Old High German poem about the end of the world and in the versified Old Saxon life of Jesus (Heliand “The Savior”). Countless works have been written about this word, but the most reasonable interpretation is probably the simplest one: –spill means “destroy”, and mu– seems to designate “earth.” That being said, one certainly notices numerous mo ~ mu- words that refer to wetness, but they do not exhaust the picture. That is why above, I proposed the gloss “loose earth.” Sand and all kinds of grain can also be subsumed under the concept of looseness, and dirt ~ mud belong here too.

Muspell and the end of the world. 1908, from Norse Stories Retold from the Eddas by Hamilton Wright Mabie. Public domain via Wikimedia Commons.

Now the ghost of the s-mobile will appear before us again and refuse to be exorcised. Engl. smut, which we remember only as meaning “filthy language,” turned up in English texts in the seventeenth century; it designated “black, sooty mark.” Od English had smitt “to smear,” smittian “pollute,” and other words of this type. Even smite once meant “to defile,” that is, probably “to throw, sling mud at.” Smudge, synonymous with the now forgotten smutch, is another newcomer (no records before the seventeenth and even the eighteenth century). It ends in the expressive –dge sound. Surprisingly, German Schmutz “dirt” also turned up in texts only in the fifteenth century, that is, very late. If the original root was mu-, did s- append itself to it for the sake of emphasis? Or did people coin many sound-symbolic sm-words and later “abstracted” mu– from them? Chronology seems to speak against this hypothesis.

In the post for August 21, 2019, on the word smell, I discussed a few sm– words. The list is long, especially if we look at the vocabulary of various languages, but, since sm– is not limited to “muddy words,” smell does not belong to our story today (except for reminding us that language tends to be expressive), it is reasonable to stay with a more limited group. Mud is certainly a cognate of smut ~ Schmutz. If the root mu– “earth; fragmented earth; loose earth; wet earth” existed, we witness the most “primitive” process of word formation that can be imagined. People probably coin the same words again and again. The ancient layer would disappear and be replaced by its twin. Nothing can be more primitive than the syllable mu, but why was it associated with “earth,” and why was it reinforced with initial s, so that smu– began to compete with mu-? We have little chance of answering these questions. But it becomes clear why linguists, both perfectly sane and deluded, have more than once tried to reconstruct a limited number of the first (very first!) roots, from which the entire vocabulary of all the languages could be derived. Rosh, yon? Mo, mu? It is better not to chase this rainbow. Mud is a primitive substance, and the word designating it also seems to be “primitive.” From mud to mucus, and all the way back to mu– in Muspell? This is a journey worth taking, for at the end of it we probably come close to the secret of the origins of words.

Here’s mud in your eye. A toast by August Hermann Knoop. Public domain via Wikimedia Commons.

Featured image credit: Mud via mnplatypus. Public domain via Pixabay. Image has been cropped to fit.

Recent Comments

  1. Constantinos Ragazas


    “A non-Germanic Indo-European related verb [to “hunt”], if it exists, should begin with k, and Greek kíein “to go, move,” which, naturally, occurred to our earliest etymologists, especially because it is the root of the Greek verb meaning “to hunt,” has a root vowel incompatible with short u in Old English.”

    Happy to hear you say that! You could have said that in your previous post. But you didn’t.

    The Greek for “hunt” is “κυνηγι”. What “root vowel incompatible with short u” are you thinking of? The Greek “υ” in “κυνηγι” [“hunt”] often is written as the Latin “u”. As, for example, with “super” from Greek “υπερ”.

    “In early Greek … [“υ”] was pronounced [u]. ”


  2. P. Taylor

    It is worth noting the discussion of the Samoyedic words for “earth” (Nganasan моу, etc.) in Ante Aikio (2002), “New and Old Samoyed Etymologies”, especially pages 22-23. The Samoyedic words invite comparison to Finnish muta “mud”, Estonian muda “‘Kot, Schlamm, Meerauswurf'”, Northern Sami mođđi “mud” (with further cognates in other Samic languages), and Moksha мода “land, field, soil” and Erzya мода “soil”. When taken together, these cognates indicate that the word is old in Uralic and suggest the reconstruction of a Proto-Uralic *muďa. Aikio also proposes that the Germanic words, in turn, ultimately reflect a very old borrowing from Uralic.

  3. Yves Rehbein

    I was just going to mention Michael Peyrot (2019) “Indo-Uralic, Indo-Anatolian, Indo-Tocharian”, in which, incidently, Aikio is also mentioned.

    I meant to mention it for another reason though, because “Next to the widespread PIE interrogative stem *kwi-, *kwe-, *kwo-, there is another interrogative in *m- …”.

    As for the criticism, the author following Aikio summarizes a nearly exhaustive argument to show that one root (“to drink”) is not reconstructable for both language families. I think it’s fair they point that out, while the book which the chapter is part of tends tob support the Indo-Uralic hypothesis though. They do reservedly not say it were a loan instead. Personally, given the supposed time depths, I find it undecidable, and that eventually both the discussed reconstructions (three, including PIE, which isn’t in doubt) might be valid.

    I should wonder whether that’s still on-topic. On the one hand, it puts things into perspective, but hardly matters for a blog mostly concerned with English.

    On the other hand, it’s still worthwhile if mentioned “drink” leads us onto foodstuff. I mean, “Matsch” is much more frequent than Low German “Modder” (while “Moder” only anymore appears as stem in the verb “vermodern”). The former can well describe food of a certain make, I think, also childish “Matschepampe”; compare “mashed potatoes” (Quetschkartoffeln”, “Kartoffelbrei”), “mush”, “mousse”, Ger *Maische*, if only for an unetymologic comparison. This perspective opens several more comparisons. If I think of a certain root (to knead, viz “mix”, “make”, though I doubt the latter a little), then I also have to consider *dough*, *dʰéyǵʰ- “to knead”, and indeed *dʰéǵʰōm (e.g. Slavic “zama”), *ǵʰmō- “earth” readily compare. The link may be seen in clay, pottery, compare German “Ton” (clay).

    But, this explains “mud” not to the full extend. Considering ominous *m- above might have a lot more to say. Just compare *sem- ~ *sm-, “some”, and “so many” (head node to Kostas bringing up pronouns as potential source of the s-mobile once).

  4. Constantinos Ragazas


    Upon further reflection, more likely the root for the Greek “κυνηγι” (“hunt”) is the ancient Greek “κυνος” (“hound”). Or both may derive from the Greek kíein “to go, move,”.

    And that makes sense. Since “hunt” involves the use of “hounds” in the chase. While “hound” likely derives from “hunt”. This association makes the Greek “κυνηγι” even more likely the origin for the Engl. “hunt”.


  5. Constantinos Ragazas

    As for “mud” (water or not), it may be relevant that the Greek for “soil” is “χωμα”. The “mu”/”μα” in “mud” and “χωμα” often appears in other Greek words. With some association to ‘material being or thing’. Probably connected to “mother” and “matter”.


  6. Constantinos Ragazas

    Hello Yves Rehbein,

    “…head node to Kostas bringing up pronouns as potential source of the s-mobile once”.

    Thank you for that.

    But also, I like to suggest that words from a polysyllabic language (eg Greek) can often contract to less syllables in a less syllabic language (eg English).

    I have many examples of that. For instance: “tree” from Gk “δεντρο”; “grass” from Gk “γρασσιδι”; “chair” from Gk “καρεκλα”; “pain” from Gk “πονος”; “name” from Gk “ονομα”; “gargle” from Gk “γαργαλα”; “cry” from Gk “κραυγη”; “frog” from Gk “βατραχι”. Etc.

    This observation should be considered when considering the etymology of words with “source unknown”.


  7. Fred de Vries

    Old English had smitt “to smear,” but modern Dutch has smerig “dirty” and smeer “grease”.

  8. Bernard Leopold

    I’m trying to track down the origin of a family name from Pomerania: Möde. In the US it usually take the variant Moede. I have also seen the variant Moet, esp. Dutch Van der Moet. In Lower Saxony there is an area called Moide near the Lüneburger Heide. These areas are all lowlands, with numerous lakes and bogs. Do you think these place- and family-names have the same “mu-” root you are discussing here?

Leave a Comment

Your email address will not be published. Required fields are marked *