As something of a follow-on to my question about “naturalizing” conlangs, I’m looking for any information - actual tools would be nice, too, if available - for “borrowing” from one conlang into another in a realistic manner. There would be a number of subsidiary questions for this, some of which I include here:

  • Assuming that I’m borrowing from sourcelang to destlang, should I destlangize the word, or leave it in its “native” form (ignoring issues of orthography relative to different writing systems)? By destlangize, I mean things like…

    • Does the borrowed word take on the conjugation/declension pattern of destlang?
    • Does the borrowed word get re-spelled to conform to destlang’s orthographical conventions?
    • Does the borrowed word’s pronunciation get changed to conform to destlang’s phonemic conventions (e.g., elimination of consonant blends, all syllables must end in a vowel, etc.)?
  • What sort of methods exist for choosing which words to borrow?

  • What causes a “native” word to fall out of use in favor of a borrowing, or, contrariwise, why would both the “native” word and the borrowed word stay in use?

(At the suggestion of a commenter, I wish to make it clear that I am interested in this issue primarily with respect to “naturalistic” conlangs and natlangs.)

  • Too short to be an answer, but if the more prestigious sourcelang is to destlang, the more of the pronun/writing is kept (as a rule of thumb).
    – Duncan
    Commented Jun 6, 2018 at 18:06

5 Answers 5


Should I destlangize the word, or leave it in its “native” form?

In natural languages, borrowed words are almost always "destlang-ized" to some degree, but it won't necessarily always to the same degree. Even within the same language, often some borrowed words will be more integrated into the language than others. The more recently a word was borrowed, the more speakers of the language will treat it as a "foreign" word and be willing to make exceptions for it, but as its use becomes more and more common, speakers will inevitably begin to treat it as a native word if it doesn't fall out of use.

Does the borrowed word take on the conjugation/declension pattern of destlang?

Sometimes, sometimes not. There are examples of both in natlangs.

The English verb "to google" has been borrowed into many languages, and it seems to typically conform to the conjugation patterns of those languages. Even in Japanese, a language with a famously closed class of verbs, borrowed "to google" as a verb ググる(it may have become a verb after being borrowed as a noun, but either way, it's quite remarkable).

Yet it's also possible for exceptions to be made. This happens with some English noun borrowings (most English speakers won't try to pluralize edamame or Pokémon with English plural morphology), but not with others (many English speakers will pluralize zucchini as zucchinis, and you'll pretty much never heard zucchino).

In my experience, verbs are more likely to take on the native conjugation patterns than nouns are native declension patterns, but I don't have any data to back that up cross-linguistically. A borrowed word is definitely more likely to be conjugated or declined like a native one the longer it has been in the language (the pedantry about pluralizing Latin borrowings in English is an exception to this, but that's more because of the social prestige associated with having studied Latin).

Does the borrowed word get re-spelled to conform to destlang’s orthographical conventions?

This really is a case-by-case thing. If the language uses a completely different writing system, almost definitely, but sometimes the orthographical conventions won't be the same as native words. Japanese uses katakana for borrowings, for instance, and English borrowings from Japanese and Chinese tend to follow whatever romanization scheme was most popular when they were borrowed, even if the pronunciation is pretty opaque. For instance, the Chinese word 道 was borrowed as "Tao" in English based on older Chinese romanizations, even though the Chinese word now romanized as dào and the English derived word "Taoism" is more often pronounced /ˈdaʊ.ɪzəm/ in English (though a spelling pronunciation /ˈtaʊ.ɪzəm/ has arisen due to this).

Does the borrowed word’s pronunciation get changed to conform to destlang’s phonemic conventions (e.g., elimination of consonant blends, all syllables must end in a vowel, etc.)?

Almost always yes. While sometimes educated speakers or upper-class speakers will try to pronounce a borrowed word with its "original pronunciation" rather than its borrowed one ("gyro" is a good example of this), even those pronunciations are almost always somewhat changed to fit the destlang's phonology better, and often they aren't even closer to the pronunciation in the sourcelang. How exactly the word is altered to fit the destlang's phonology depends on the particulars of the borrowing situation, however.

What sort of methods exist for choosing which words to borrow? What causes a “native” word to fall out of use in favor of a borrowing, or, contrariwise, why would both the “native” word and the borrowed word stay in use?

I've combined these questions because I feel there's a lot of overlap in the answers. Often, words are borrowed to fill a semantic gap in the destlang, which is why words for new technologies are so often borrowed (think of how many languages borrowed the word "television"!)

Borrowings are also often used to refer to the "versions" of certain things from the part of the world that speaks that language -- consider how anime, a generic word for all animation in Japanese, was borrowed into English as a word for a style of Japanese animation, or how English borrowed "chai" from the Hindi/Urdu word for tea, चाय (cāy) / چای‎ (ćāy) to refer to tea with certain spices based on Indian recipes.

Also, sometimes words are borrowed because they carry some social value seen attractive by the speakers of destlang. Maybe destlang speakers think sourcelang sounds refined and sophisticated, and so upper-class destlang speakers borrow words from sourcelang to sound fancy. Of course, this sort of situation depends on a lot of social and sociopolitical factors wherever the languages are spoken.

  • 1
    A very complete answer! Nicely done!
    – elemtilas
    Commented May 23, 2018 at 23:42
  • "Pokemon" not getting a "normal" English plural is easily explained by the fact there are so many examples in English of existing words that don't do it either: in my dialect of English, deer, fish, moose, elk, buffalo, caribou and other such words don't change between singular and plural. "Pokemon" is treated the same way. In a language where the rules for distinguishing between singular and plural are stricter, that might not be the case. Commented Jun 12, 2018 at 17:01
  • @KeithMorrison Despite English's number of irregular plurals, native English-speaking children still consistently apply its regular plural morphology to nonsense words (see: the Wug test). The fact that English allows irregular plurals similar to this no doubt contributes, but without any influence from its native morphology I highly doubt it would be assigned this irregular plural morphology. But yeah, it's not necessarily going to carry over that way in all cases in every language!
    – Sparksbet
    Commented Jun 12, 2018 at 17:51
  • @Sparksbet, the sourcelang morphology has an influence, sure, but what I'm saying is that English finds it easy to accept identical singular/plural forms because there's already so many examples, especially for words for animals (which Pokémon could be categorized as). On the other hand, in Turkish, the plural of "Pokémon" is "Pokémonlar", exactly what one would expect the plural in Turkish to be. Commented Jun 20, 2018 at 15:20
  • 1
    @KeithMorrison I'm simply using Pokémon as one of several examples in which sometimes sourcelang morphology influences destlang morphology - of course not every language will use the same morphology for loanwords from the same source, and in the case of very recent loans like "Pokémon", there may even be differences between speakers (my grandmother definitely says "Pokémons"!) But I think the Japanese examples serve as well as any when it comes to the influence of sourcelang morphology; perhaps I'll include some Latin loan plurals as additional examples when I have time to edit.
    – Sparksbet
    Commented Jun 20, 2018 at 16:10

There are a few words in Esperanto that are arguably borrowed from Ido; examples include olda "maljuna, malnova" and kurta "mallonga". With respect to the subquestions: These words didn't need any changes, they already blend perfectly into Esperanto. The borrowings from Ido still are in a niche position in Esperanto and live mainly in the poetic register.

Of course Ido (as a fork of Esperanto) shares a lot of vocabulary with Esperanto, but this is due to inheritance and not borrowing.

I also remember vaguely that some of Tolkien's Elvish words were borrowings between Sindarin and Quenya, but unfortunately I don't have an example at hand. This would be a "constructed borrowing" between conlangs, adding to their naturalness and their diachronic depth.

  • Esperanto and Ido are somewhat outside the area of interest/intent of this question; neither language does - or was intended to - seem like a 'natural' language - they are most definitely, by design, artificial languages specifically intended as interlanguages. The Quenya/Edhellen cross-borrowing would fit into the area of interest/intent, but I'm less interested in extant examples than I am in process. Commented May 22, 2018 at 16:26
  • 1
    @JeffZeitlin you might consider specifying in your question that you're interested in how borrowing works in natlangs and naturalistic conlangs, if that's specifically what you're interested in.
    – Sparksbet
    Commented May 22, 2018 at 20:23
  • @Sparksbet - A good point; I suppose I assumed that that would be understood since I explicitly mentioned my previous question about 'naturalizing' conlangs, and indicated that I considered this a 'follow-on'. Commented May 23, 2018 at 11:21

One thing you may want to consider is how the sourcelang word reached the destlang. If the word was spread through literature to the destlang speakers, then it may be spelled the same as it is in the sourcelang but pronounced according to the pronunciation of the destlang. If the word is spread through oral communication then it may sound like the sourcelang but be spelled differently. One example would be ray (or rai), a Spanish word descended from the English ride as in I got a ride home.

Another thing you may want to consider is the possibility of usages that sourcelang speakers may consider "incorrect". An example of this would be the French word le parking, which means car park in English. In the case of "incorrect" usage the important thing to consider is which social societal group naturalized the word and in what context. If the word was naturalized by academics, then it may continue to be pronounced and used the same way it is in the sourcelang (as in the cases of many Latin words used in English). If it was naturalized by non-linguists in normal social contexts, then it may be naturalized with spelling errors, mispronunciations, and odd usages (consider le shopping in every day French).


An addition to Sparksbet's excellent answer I'd add one tool to your armamentarium:


Sometimes foreign words borrowed a long time ago from L1 become naturalised in L2 only to become denaturalised again in later times. In other words, the legitimately L2 word becomes more like the antecedent L1 word.

Case in point, PEKING. English borrowed this city name centuries ago and happily naturalised it (as we tend to do sooner or later with every word we borrow!). But recently, you've probably noticed an odd spelling "BEIJING". For some strange reason, this word has de-evolved from its naturalised form to something approaching a modern Mandarin pronunciation-spelling of the name, rather than the English pronunciation of the name.

This doesn't always happen. Thus far, we continue to pronounce Baile Átha Cliath as DUBLIN; Moskva as MOSCOW; and Krung Thep Maha Nakhon as BANGKOK.

  • 4
    Peking was borrowed based on older romanizations of the word. Beijing is based on modern pinyin romanization, and it too has been nativized to some extent (replacing t͡ɕ with d͡ʒ, for instance). Referring to that as "de-evolution" is problematic. Rather, it was re-borrowed because of sociopolitical factors as China regained international prominence.
    – Sparksbet
    Commented May 24, 2018 at 0:18
  • 1
    De-evolution with respect to English: the former nativised form has been replaced with a newer foreign form. No worries, though! It too shall be assimilated. After all, that's what English does best! (We can see this happening, for example, with the increasing use of "beijing duck", rather than the older "peking duck".)
    – elemtilas
    Commented May 24, 2018 at 0:37
  • 1
    "Peking" really isn't any more nativized just because it's based on an older spelling pronunciation than "beijing" is. De-evolution really isn't a good word to use here -- it carries negative connotations in addition to not really being an accurate description.
    – Sparksbet
    Commented May 24, 2018 at 0:56
  • 2
    I think there are better examples for this process, I can think of the words debt and doubt that acquired their b's due to it.
    – Sir Cornflakes
    Commented May 24, 2018 at 8:33
  • @Sparksbet Thank you for your opinion! We don't agree on the negative connotation aspect, but as far as a mechanism of borrowing goes, I think it stands as well as any of the others mentioned.
    – elemtilas
    Commented May 24, 2018 at 15:42

Its common for languages to borrow the words for trade goods introduced to them by other cultures. These are called 'Wanderwoerter', or 'wandering words' in German. All the words for 'tea' in our world, for instance, can be traced back to just two distinct roots. One gave rise to the word for tea in China and Mongolia, the other is the common ancestor for the words for 'tea' everywhere else. Words for region-specific materials also tend to get borrowed. Though they can also just be named after foreign locations that the people who speak the destlang got the material from (we have quite a number of these in English).

Words for technologies also tend to migrate. Hungarian has multiple words related to horse riding that originated in Turkish.

Also note how English likes to borrow foreign words for foreign objects. We have words like sombrero from Spanish and kimono from Japanese. Note however that borrowed words may not hold the same meaning as they do in their parent languages. In Spanish the word 'sombrero' is the word for any hat with a brim. And kimono is the name for a specific garment, but in English we normally use it to refer to any kind of traditional Japanese dress. This is pretty common cross-linguistically.

Also note that some languages borrow foreign words more than others. Some prefer to just derive new words from native words. Finnish used to be like this. German still is to a large extent (though that doesn't mean loan words don't exist in German, there are a few from French and English, such as bon-bon, pommes frites, and jeans).

  • Deriving new vocabulary from existing words instead of simply borrowing them outright (well, mugging other languages, in the case of English) can also depend on official government policy. Iceland does it deliberately, and generally successfully. France has the old fogies club of L'Académie Française and backing of the government who do it somewhat less successfully. Other languages and countries, of course, don't even bother trying. Commented Jun 25, 2018 at 16:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.