This is something that I have been wondering about for a while, and I thought that I could ask about it here. I am unsure about whether this is strictly "on-topic" because it may be only half about the English language. I think that it is ok because it is mainly asking about how a word came into English.

My question is about the word "lakh", which is oftentimes used in India (and Pakistan/Sri Lanka/ Bangladesh, too) to replace one hundred thousand. (100,000)

I looked in some dictionaries, which include this word and its definition. They also say that it is an "Indian" word.

Some of the places I checked, like Wikipedia, say that the word "lakh" comes from the Sanskrit word "laksha." I am fairly certain that this is true, as I have read about this before.

In my native language Malayalam, the word is "laksham." I also know that the word is very similar in Kannada, but it doesn't have an "m" sound at the end. The original word had a "ksh" sound, but in the English word, it only has a "kh" sound, and that "sh" part is lost.

There is also a Union Territory in India called "Lakshadweep", which literally means "one lakh islands." I looked at the origin of the word, and it does in fact come from Sanskrit.

So why is this? Why did the "sh" sound get lost? Did the first Britishers to come to India mishear the word, and did this mistake sort of "stick"? Or is it because of some rule that I do not know about?

I would appreciate an answer and thank you in advance.

  • 4
    The Wikipedia page "Lakh" is terrible, very poorly sourced and badly written. Some of its references are 61, 90, and 122 years old, some are by people with no skill at linguistics/philology/lexicography (a book on insects!), and there is often disconnect between the source and the article (it's unclear which part of each sentence is actually included in the cited source). A good modern dictionary or academic text produced by experts with the latest contemporary knowledge and methodology will explain the actual origin far better.
    – Stuart F
    Mar 9, 2023 at 16:10

1 Answer 1


This numbering system is a fairly well-known peculiarity of Indian English, but isn’t actually much of a mystery; the answer to this question is pretty straightforward. Oxford Languages gives the origin of the word ‘lakh’ as simply:

via Hindi from Sanskrit lakṣa

In other words, the word wasn’t anglicized from Sanskrit lakṣa, but from the Hindi lākh.

The /ʃ/ sound (sh) was lost somewhere in the Sanskrit ⇾ Hindi conversion. This is similar to how Hindi rakhna (राखनं) is different from the Sanskrit word it comes from, rakshati (रक्षति). The loss of the final /a/ sound is natural, as with akṣara becoming akṣar in Hindi.

There's more information about this shift here, as @Au101 notes in a comment. The relevant quote is that "A sibilant assimilates to a preceding k as an aspirate (i.e., the sequence kṣ becomes kkh)."

As you mentioned, the original Sanskrit word lakṣa (laksha, लक्ष) has been adopted into many different Indian languages. Wikipedia has a list of how the word appears in several languages:

  • In Assamese: লক্ষ lokhyo, or লাখ lakh
  • In Bengali: natively (tadbhava) known as লাখ lākh, though some use the ardha-tatsama লক্ষ lokkho
  • In Hindi: लाख lākh
  • In Dhivehi: ލައްކަ la'kha
  • In Gujarati: લાખ lākh
  • In Kannada: ಲಕ್ಷ lakṣha
  • In Kashmiri: لَچھ lachh
  • In Khasi: lak
  • In Malayalam: ലക്ഷം laksham
  • In Marathi: लाख / lākh/laksha
  • In Meitei: ꯂꯥꯛ lāk
  • In Nepali: लाख lākh
  • In Odia: ଲକ୍ଷ lôkhyô
  • In Punjabi: (Shahmukhi: لکھ, Gurmukhi: ਲੱਖ) lakkh
  • In Sinhala: ලක්ෂ lakṣa
  • In Tamil: இலட்சம் latcham
  • In Telugu: లక్ష lakṣha
  • In Urdu: لاکھ lākh

In these, you’ll notice that the languages that preserved the -sh- sound in the middle of the word are all from the Southern parts of India (Kannada, Marathi*, Malayalam, Sinhala*, Tamil, Telugu). Kashmiri’s لَچھ is a bit of an outlier, as it pronounces it with a ‘ch’ sound. (I can’t explain it.) Notice how the /k/ sound also becomes aspirated as we move up the map.

When the British first began to try to write down and understand Indian languages, they would have been interacting with Northerners, such as speakers of Hindi or Punjabi. Naturally, this is the spelling they would have recorded.

(The rest of this answer is just a whole lot of fluff that came from my venturing down this rabbit hole. No need to read past this.)

In early texts, there were some alternative spellings for lakṣa, as an entry for from A Comparative Dictionary of the Indo-Aryan Languages notes:

lakṣá: 1] stake, prize 2] mark, sign 3] 100,000 [As numeral, see Add.]

Addenda: lakṣá: lákkh ʻ100,000ʼ, lākha, lakha, lakka

(The original entry was confusing, I've edited the formatting a bit)

Other spellings I’ve found include, from John Shakespear’s 1845 An introduction to the Hindustani Language, lāk’h; and from John Borthwick Gilchrist’s 1796 A_Grammar_of_the_Hindoostanee_Language, lak,h. There are also two spellings in this quote from The popular Encyclopedia (1817):

LAC, or Lak, from the Sanskrit lakshâ or laksha, that is, 100,000. In the East Indies it is applied to the computation of money. Thus, a lac of rupees is 100,000.

A few other spellings here, from The Concise Dictionary of English Etymology:

Lac (2), a hundred thousand. (Hind.- Skt.) A lac of rupees = 100,000 rupees. Hindustani lak (also la̍kh), a lac. - Skt. laksha, a hundred thousand

And there’s this somewhat bizarre definition from A Concise Etymological Dictionary of the Dhivehi Language:

LAKKA [?]: n. & adj. Hundred thousand; many, numberless, incalculable, myriad, several. infinite. innumeralble. [sic] Sin. laksha: lac (100,000). Ar. lakk. Alāk. lukūk: lac.
Sk. laksha: a lac, one hundred thousand. P. lakkha: a high number; a lac or 100,000. (Cf. laksha).
LAKSH’A [?]: n. & adj. (arch) Hundred thousand; many, numberless, incalculable, myriad, several, infinite, innumerable. Sin. laksha: lac, 100,000

(Dhivehi is spoken in the Maldives, but this does suggest that there were a few other spellings floating around at the time)

Of these, the most competitive spelling was lac, as this excerpt from the East India Company's 1813 Charter Act (section 43) shows:

a sum of not less than one lac of rupees in each year shall be set apart and applied to the revival and improvement...

But the other spellings have died out; today lakh is the only accepted one.

There is more to see; for one, consider this quote from the 1812 A_Grammar_of_the_Malayan_Language:

laksa: one hundred thousand...In borrowing the word laksa from the Sanskrit, the Malays have, very arbitrarily, changed its signification from an hundred thousand to ten thousand ; which often gives rise to misunderstandings in their transactions with merchants from the continent of India.

There are three implications here- the first is that laksa has a different meaning in Malay, the second is that ‘signfication’ [?] is a word, and most importantly, that Indian merchants did hear and understand the word ‘laksa’ in the context of 100,000, even if their own native languages had changed the spelling and pronunciation.

Finally: this excerpt from 1764 Ames Library Pamphlet, in French, uses the ‘lakh’ spelling- this implies that ‘lakh’ was already the established orthographic precedent. And, ‘lakh’ was chosen over another alternative spelling, lac, which would have been more naturally French. Note that other Indian words, like ‘roupies’ (rupees) were gallicized.

[?] ready money » a contribué pour un lakh de roupies à la construction des nouveau bâtiments de cette université. La meme philanthrope a offert un prix de cinq mille roupies au parsi qui se distinguera comme avocat plaidant (barrister) a la haute cour de Bombay.

As for the name of Lakshadweep, Byju's has this to say:

This set of small islands was earlier known as Laccadive, Minicoy and Amindive islands. In 1973, it got changed to a collective name, Lakshadweep.

Likely, the Parliament in India decided to keep with the original Sanskrit root. I suspect that this is because it’s a Southern state, but I’m really not sure.

The font here can't completely support 'a̍' and 'ṣ,' so sorry that it looks a little strange

  • 2
    FWIW simplification of Sanskrit consonant clusters (like kṣ) is an extremely well-established process in the history of Indian languages. Obviously there's a limit to the scope of this Q&A, I don't wanna lead you too far down rabbit holes, but as we transition from Sanskrit to modern Indian languages we pass through the Middle Indo-Aryan Languages (often called 'Prakrits'). Pali is one of these and is the language of Buddhism's Pali canon. This is a great resource ancient-buddhist-texts.net/Textual-Studies/Grammar/… that shows how in Pali kṣ -> kkh
    – Au101
    Mar 9, 2023 at 17:25
  • 2
    A similar example you might be more familiar with is Sanskrit nirvāṇa (we've all heard that word, right?) In Pali rv simplifies to vv, and vv becomes bb giving the Pali version nibbāna which our Buddhist friends might have come across before
    – Au101
    Mar 9, 2023 at 17:28
  • @Au101 - Fascinating. I've edited my answer to add that link and I include some of that information, although not all of it. Thank you for that :) Mar 9, 2023 at 20:07
  • 1
    Some of those languages are not only from the South but also not at all Indo-European, so not (not even indirect) descendents of Sanskṛt, so they only borrowed the word. Mar 11, 2023 at 21:08

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.