Does Swahili (and hence cognates of other Bantu languages) simba have any relation to सिंह (~singh or ~simha in Hindi and Sanskrit respectively according to Wiktionary, please correct me if I am wrong), or is it simply a coincidence? After all, they sound very similar and have almost identical meanings. If so, did this borrowing result from contact? Did Bantu and Indic/Indo-European language families have common ancestors? A cursory search on Wikipedia doesn't yield any clues.

Links: https://en.wiktionary.org/wiki/Reconstruction:Proto-Bantu/%C7%B9c%C3%ADmb%C3%A1, https://en.wiktionary.org/wiki/simba#Swahili, https://en.wiktionary.org/wiki/सिंह) https://en.wikipedia.org/wiki/Indo-European_languages#Suggested_macrofamilies


6 Answers 6


That is a coincidence, the two words are not related, neither are the Indo-European and Bantu languages.

The Swahili simba 'lion' comes from the Proto-Bantu *ǹcímbá 'any of various wild felines or similar, including wildcat, lion, leopard, civet, genet'.

  • 4
    @Anonymous - Yes, there are really many coincidences in unrelated languages.
    – Yellow Sky
    Nov 5, 2017 at 17:40
  • 5
    Coincidences like this are very common. Any two languages taken at random will divulge a dozen or two words or word-like chunks that look or sound very similar and mean very similar things. That's unavoidable because there are many orders of magnitude fewer word-shapes available in any language than there are meanings to apply to them. The devil is in the details, in this case the meaning of "very similar" -- just how similar do meanings and sounds have to be?
    – jlawler
    Nov 5, 2017 at 21:11
  • 5
    The majority view among Indologists is that the Sanskrit siṁha “lion” does not have an Indo-European etymology, but that it is a substrate word. Thus, the question of whether or not Indo-European and Bantu are related is of no relevance to this question.
    – fdb
    Nov 5, 2017 at 23:18
  • 1
    False Cognates Alert :) Nov 6, 2017 at 7:14
  • 2
    This is only half an answer, missing the other root. Wiktionary (through our @Aryaman) makes the less bold claim "is likely unrelated". The Sanskrit form is compared to Tocharian, Chinese, Tibetan--all possible to have loaned from Buddhist scripture, yes? Chechen and other languages I don't know about are mentioned, too. The thing is, a root sinĵʰás is proposed, which is obviously not the same as Proto-Bantu, however for a limited time-depth under several assumptions (e.g. about contact). I wonder how Bantu is reconstructed, if writing came very late,
    – vectory
    Jan 1, 2019 at 9:26

Indic languages are not related to the Bantu languages genetically; however, both Hindi/Urdu and Swahili were influenced by Arabic due to contact between speakers. Neither of these words arose through mutual loans from Arabic, though; 'lion' in Arabic is 'أسد,' or ' 'asada.'

  • 2
    siṁha is Sanskrit. Arabic has nothing to do with this.
    – fdb
    Nov 6, 2017 at 9:50
  • There is such a thing as borrowing. Aug 27, 2023 at 19:50

If the word is borrowed, the direction of borrowing would have to be from Bantu to Indic, given that nsimba is widely distributed in Bantu, not limited to Eastern Bantu. This is not impossible in light of cultural imports at Chanudaro from Africa around the 2nd millenium BCE (pearl millet, African women). In light of the fact that the root appears in Armenian and Vedic, the borrowing would have to date to about that period. However, the Bantu expansion would have just started at that point, and certainly would not have reached coastal East Africa (a plausible point of Afro-Indian contact). So there is no plausible point of contact between the two populations at that time.

The consonant change b→ǵʰ which would be required in light of the differences between the Indo-European words and the Bantu words is also a highly improbable change – it doesn't correspond to any sound change tendency in Indo-European.


The noun class ku- (locative, distal) would be the best fit because you don't go near the damn thing and it would support Satemization of Simba.

If this means that cognates would appear with different prefixes (?) the case of Indo-Iranian **mr̥gá- "beast" (cp. *n- animal, inanimate?) could be decisive. So let's see.

I recognize exactly none of the ca. 600 lemmas in Wiktionary's Proto-Indo-Iranian category and have within three guesses found *wŕ̥kas "wolf" and *Hŕ̥ćšas "bear", which have thorough Indo-European etymology yet a slight problem with taboo avoidance, which overshadows the lack of comparanda in several branches. The other word I looked at is *dr̥Hgʰás "long, late" after speculating about German droh "threaten", cp. *d- (abstract noun). The category is of course biased to be filled with words that correspond to the Indo-European category. I was hoping to find an exact match, because I don't quite remember if I hadn't seen one previously (citation needed).

The vocalic r can be explained as a common feature of African languages, as is said in reconstruction of Egyptian Aleph, so eg. ꜣbw is correlated to the liquid in elephant and Aramaic pīlā as well (cf. Meroitica 22; more generally Peust, I guess). Krokodil, cocodrillus also seems to match this pattern, including reduplication, though Pliny reports of an Egyptian word which is confirmed by the modern reconstruction /kɑpuː/, it has synonyms dpy mzḥ and kꜣpw.

Certqinly the situation

would also match this despite

*d- is too abstract to account for anything, but it sounds like

  • Do you want to flesh out this claim that vocalic r is common in African languages, especially in Bantu from whence simba? Africa is a big place.
    – user6726
    Jun 6, 2023 at 1:16
  • Reading this, I feel as if I have caught the last five minutes of a lecture. Vectory, would you care to explain how any of this relates to the question? The best fit for what? What can satemization possibly mean when talking about a Bantu language (if you mean "palatalization", why call it something so bizarre?) Etc.
    – Colin Fine
    Jun 6, 2023 at 15:49
  • again, I'm using different softkeyboards and sometimes hit by tab ->enter (send) on the less familiar keyboard layout, by accident, and nobI don't knowbwhere I'm going with this but but it should be sunny.
    – vectory
    Jun 6, 2023 at 21:21

Trade goes back to the 5th century BCE between India and the east coast of Africa. Under such circumstances, loan words occur between languages that are otherwise unrelated. A possible example: The word "sangha" in Sanskrit refers to "community", as does the word "sanga" in the Shona language.


i think both can be related as well as it could be a coincidence. relatedness possibilities are numerous as East African Coast of the "INDIAN" Ocean had been trading with Malabar Coast and Beyond and were connected by land since eternity. But, as Lion is one of the most prominent animal in all languages throughout the history. And Lions were found all over the world as well. Therefore introduction of the word, to and from India (or Indo-european) and Africa (sawahili) origins is equally plausible.
On the other hand, convergent evolution in human societies is a common occurrence. seemingly independent groups of people developed similar tools, skills, social structures and designs. similarly many words in unrelated languages often sound similar because ultimately all words originated from sounds. e.g. words describing sounds, animal sounds and natural phenomenological sounds and sounds associated with basic human actions and emotions like yawning, coughing, sobbing, laughing, etc. Thus both convergent development or built-in tendencies or genetically determined human social and intellectual progress resulting in similarities between unrelated societal groups should be explored more. And at the same time we should be open minded in examining the possibility of complex interactions and relations among distant societies. Trans-Atlantic Pre-Columbian interactions have recently been discussed, similarly ancient Indo-African interaction and diffusion should also be explored further.

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