How many languages can a person learn in their lifetime?

[vonPeterhof]

The last Polyglot Gathering actually included a lecture about research into the the phenomenon of polyglottery, and one of the topics discussed was the existence of a limit to the number of languages one can learn in a lifetime (unfortunately I can't link to a specific time in the YouTube link below, but the relevant part starts around 16:37). To summarize, while existing research hasn't yet revealed a biological limit (aside from the human lifespan itself), interviews with polyglots who can speak dozens of languages with varying degrees of proficiency (conducted by Erik V. Gunnemark and Dmitri L. Spivak) reveal that the number of languages they believe they know "fluently, deeply and completely" tends to be 7 ± 2. Whether or not this is actually related to the capacity of the human working memory cannot be stated without further research.

[PeterMollenburg]

Give or take 1000 hours per language (accounting for personal study time too) at 40 hours per week, that's two languages per year. Starting at 18 (you're in school before then!) and "retiring" at 70 that gives you 52 years.

So 104 languages if you make a full time career out of it.

I can't find any claims that anyone has ever reached three digits: Mezzofanti knew 39, Ziad Fazah claims 59, Kenneth Hale had "over 50". Are there any higher claims?

This analysis is (perhaps deliberately?) completely devoid of any time required to maintain languages, which, if taken into consideration, would reduce this calculated final number massively.

[Systematiker]

I’ve voted in the “over 20” category, though I’d hesitate to think it need be as large as it is. Here’s my hypothetical why:

Let’s take an individual and start with a foundation. English, a Romance language (I think best Spanish, but any of the major ones), and German. Let’s raise them bilingually in English and e.g. Spanish, and put them on a trajectory to learn C2+ German. Let’s even have them go to Uni in Germany. With exceptional skills in those three languages, one could pick up the following to B2 (or perhaps even extrapolate at a B2 level on the spot!) with very, very little effort:
Scots, Swedish, Danish, Norwegian, Dutch, Platt, Swabian, Bavarian (etc, I’ll stop with those three); Portuguese, Galician, French, Catalan, Occitan, Italian. A bit more effort (or I’m not sure due to resource availability or my inexperience): Latin, Romanian, Romansh, and a number of other majority Romance languages.

You’ll see I’ve named 20 total, with a goodly number of minority Germanic and minority Romance languages still out there.

There’s of course a real limit on time and maintenance, and the restrictions given by the availability of materials to “keep up” the minority languages are an issue, but with the synergy, it could probably even be done in a concentrated decade or so. Certainly by the time our hypothetical individual is 35.

But what would be the point? It’s just a min-max for numbers (Monty haul anyone?), and it’s a rather unlikely situation to be able to do it. And it would be awfully hard to measure, since many of these languages are so close, what might really be an error (wrong language!) works out fine as B2 in the language to be measured.

[galaxyrocker]

I've probably reached my personal limit for high level languages. It takes a lot of effort and sacrifice to keep them at a high level. Leosmith has written before about B2 being the "sweet-spot" for languages in that a language at this level doesn't need to be studied but merely maintained.

I think this is the key. If you're just getting a language to B2, taking a test, and then forgetting it, you could do quite a bit more than if you needed to maintain all of them at B2 for X amount of time.

Scots, Swedish, Danish, Norwegian, Dutch, Platt, Swabian, Bavarian (etc, I’ll stop with those three); Portuguese, Galician, French, Catalan, Occitan, Italian. A bit more effort (or I’m not sure due to resource availability or my inexperience): Latin, Romanian, Romansh, and a number of other majority Romance languages.

Yep. It also depends on how you count languages and what types you're looking at. I'd say if the original criteria was more specific (i.e. 'of separate families'), you could get better results. But right now you can pick up other ones in the same family, and it'll come a whole lot quicker.

[Dylan95]

Well it depends. Are they choosing a bunch of unrelated languages? Or are they choosing all the romance and germanic languages?

[leosmith]

So in the title you say "a person" and then you say

Thanks. Fixed, and you can change your vote if you want.

[leosmith]

Maybe the most impressive multi-language demonstration video I've seen is this one with Emanuele Marini speaking 16 languages at the Polyglot Conference in Budapest in 2013.

Great to see all these (mostly) European languages. It would be nice to see a few asian languages in the mix too.

[Axon]

I'm a little surprised to see that (as of now) I'm only the second person to vote 50-100. There's no way *I'm* getting to 50, but I think it's absolutely possible and has probably been done.

I believe the 50+ language accounts about Ken Hale, Emil Krebs, George Sauerwein, Harold Williams, and others. For someone to be interested in languages from a young age until their seventies or eighties gives them plenty of time to learn fifty languages. I also believe that there are some people (probably these folks) who are able to compartmentalize their languages in different ways than I.

Knowing English, German, and some Danish and Dutch, I can puzzle my way through a Norwegian nonfiction or news article. But it doesn't seem far-fetched at all that someone else with my same background could not only read that article but immediately *use those Norwegian words*. That is, transferring passive knowledge to active knowledge far more rapidly than I can.

[PeterMollenburg]

I'm a little surprised to see that (as of now) I'm only the second person to vote 50-100. There's no way *I'm* getting to 50, but I think it's absolutely possible and has probably been done.

I believe the 50+ language accounts about Ken Hale, Emil Krebs, George Sauerwein, Harold Williams, and others. For someone to be interested in languages from a young age until their seventies or eighties gives them plenty of time to learn fifty languages. I also believe that there are some people (probably these folks) who are able to compartmentalize their languages in different ways than I.

Knowing English, German, and some Danish and Dutch, I can puzzle my way through a Norwegian nonfiction or news article. But it doesn't seem far-fetched at all that someone else with my same background could not only read that article but immediately *use those Norwegian words*. That is, transferring passive knowledge to active knowledge far more rapidly than I can.

Aside from depending on which languages you attempt to learn, the list is vastly cut back when you take into consideration language maintenance and what level we are talking about. It's certainly a MASSIVE underaking to take a language to C1/C2, and unless you've done so yourself you really underestimate the time and effort required to do so. B2 level is significantly less work, but still no walk in the park.

Edit: The original question was up to B2 level, so again I didn't read something properly!

[smallwhite]

* Depends if you get to warm up beforehand. (I think this is very important!)

* Depends if the person was told beforehand he would be tested this way, ie. whether he studied with this test in mind. (Some people don't maintain their past languages).

* Depends if you count dialects as separate languages like you do the romance ones. China, Germany and Norway seem to have a lot of mutually unintelligible dialects, while I think Malay and Indonesian are highly mutually intelligible. (This problem comes up in half of our discussions. But in fact, we don't need to limit ourselves to the official definition of a language. We can define a language as "any language/dialect/variation that takes 6 months or more to learn to speak at B2" or whatever appropriate to the discussion on hand).

(When you ask about "a person", can I answer "yes" if I think only 1 person on earth can do it, or can I answer "yes" only if I think every (applicable) person can do it?)