Historical story

How etymological research through loanwords provides insight into our linguistic history

The origin of words is central to etymological research. A relatively new field within this field of research is the study of loanwords in order to visualize historical language and culture contact. Linguist Peter Alexander Kerkhof conducts etymological research at the University of Ghent.

Linguist Peter Alexander Kerkhof:“To delve into the history of Dutch is to discover that the house you have lived in for years has an attic full of precious treasures. As an etymologist, I love to poke around this attic and see that the language we use every day carries the traces of thousands of years of human history.”

Dutch also lends itself well to this type of research:you can use loan words to see which peoples left their mark on our culture. In the Dutch vocabulary, for example, we see how the Indo-Europeans, our distant ancestors, took over agriculture from the Hunebed builders (pea). is a prehistoric loanword), how the Teutons of the Celtic Gauls learned to make iron (iron is a Celtic loanword) and how the Franks, our direct linguistic ancestors, liked the cheese of the Romans more than their own whey cheese (cheese is a loanword from Latin).

Hidden loanwords

There are also Dutch words of which only a part is borrowed from another language. An interesting group within these kinds of words are compounds of which the first part used to have the same meaning as the second part, so-called hybrid tautological compounds.

In the word 'fallow deer', for example, the Gallic word damma 'deer' is hidden. In the Middle Ages, this deer species was still simply called damme, which is closer to its Celtic origin. In the course of time, fewer and fewer people understood that this word meant a deer, and the recognizable word 'deer' was added to it for clarity - but therefore doubled. That way no one could be mistaken.

A striking number of animal names belong to this category. Below are a number of examples from Dutch and English in which the first element is a hidden loanword.

  • 'ostrich' ='ostrich' + 'bird'; struthio (Latin) ='ostrich'
  • 'turtledove' ='turtledove' + 'dove'; turtur (Latin) ='turtledove'
  • 'mule' ='mule' + 'donkey'; mulus (Latin) ='mule'
  • greyhound ' ='grey ’ + ‘hound ’; grey (Norwegian) ='bitch'

Germanic homonyms

The example of the greyhound shows that it is not necessarily about compositions with borrowed language material, but rather about an obscure vocabulary. Both parts of the compound come from Germanic (Old Germanic hwind + hund) and both mean 'dog', but at some point the wind part was no longer recognized as a standalone word for dog. After all, it had become homonym with the word 'wind', which refers to the weather phenomenon. Therefore, it was supplemented with a clarifying second part. This is also the case with the Dutch words 'whale', 'sea seal' and 'tapeworm'.

  • 'greyhound' ='wind' + 'dog'; 'hwind ' (Old Germanic) ='dog'
  • 'whale' ='wal' + 'fish'; 'hwal ’ (Old Germanic) =‘whale’ (compare English ‘whale ’)
  • 'sea seal' ='sea' + 'seal'; 'selha ' (Old Germanic) ='seal' (compare English 'seal ’)
  • 'tapeworm' ='ribbon' + 'worm'; 'lind ' (Old Germanic) ='worm, snake'

It doesn't always have to be animal names. The word 'meal' is a good example of a non-animal related tautological compound.

  • 'meal' ='meal' + 'time'; 'mēla ' (Old Germanic) ='time'

Composite place names

A similar kind of tautological compositions can be found in place names. In England the name is Chetwood such a case. The first part comes from Celtic (Welsh:'coed ') and means 'forest' and the second part 'wood' was added when the meaning of the first part became opaque.

In Dutch we have few such compound place names. I could find the following Dutch examples in which both elements clearly had the same meaning.

  • Monteberg (Dranouter, BE) ='monte' + 'mountain'; 'mons ' (Latin) ='mountain'.
  • Loobos (Kootwijk, NL) ='loo' + 'forest'; 'lauha ' (Old Germanic) ='forest'.
  • Waverloo (Didam, NL) ='waver' + 'loo'; 'wabro ' (Gaul) ='forest'.
  • Waverwoud (Lier, BE) ='waver' + 'woud'; 'wabro ' (Gaul) ='forest'.

Historical language confusion

That's nice to know, of course, but what's the use of this? The beauty of these compositions is that they give an indication of when a certain word was no longer understood. The fact that the composition 'turtle dove' already occurs in tenth-century Old Dutch ('turtulduva ’) makes it clear that the single word turtul had become opaque.

Tautological place names can provide interesting information about the shifting of language boundaries and the disappearance of multilingualism. In the case of Monteberg, the twelfth-century form Monteberga tells us that even then not all Flemings understood what monte meant. This may be an indication that at that time French was supplanted by Dutch in the area around Monteberg. So these words give us a valuable window into historical language confusion.

Read more from this author