How Do You Learn a New Language, Without Forgetting the Old One?

The biggest difficulty I had when learning Spanish was, perhaps surprisingly, speaking French. I had learned French to conversational fluency before learning Spanish. However, as my Spanish got better, I found it made speaking French much harder.

The problem got so bad, that after only a week or two in Spain, I could speak better Spanish than French. That’s not because my Spanish was great—I still couldn’t perform even basic functions like the past tense—but because every time I spoke French I’d stall on basic phrases and accidentally use Spanish words.

This, of course, presents a serious problem to someone like me who wants to learn multiple languages. What’s the point of learning a new language, if your previous one gets overwritten in the process?

What’s Actually Going On?

Given I’m going to be tackling four languages in a short period of time, this was definitely a problem I wanted to research. It turns out to be quite a common problem for adult learners of multiple languages.

My suspicion was that the French wasn’t lost, it was simply being dominated by my more recent Spanish. When I’d look for a word, the Spanish one came up and made it harder to retrieve the French word. Instead of being erased, my French had a layer of Spanish piled on which I had to dig through first.

Another bit of evidence with that hypothesis was that my ability to speak was impaired considerably more than my ability to understand French. This makes sense: when hearing a French word, I need to remember a connection between those exact sounds and a concept—something unrelated to my Spanish. However, when speaking, I need to go from a concept to a set of sounds, the correct responses in English and Spanish compete with French, making speaking harder.

For example, hearing the word “aussi” triggers my  memory of the word “also” in French automatically. However when I try to say the concept of “also”, I get the word in English, obviously, but also the word in Spanish “tambien” which slows my ability to retrieve the French word somewhat.

I also found that my writing ability in French wasn’t impacted nearly as much. Again, this supports the same hypothesis because writing allows longer time to reflect over which word to use. Allowing longer reaction times meant I could push past my Spanish interference and get at the correct word, which would have broken my flow in a speaking situation.

Restoring my French

This hypothesis implied that I could restore my French simply by speaking French for an extended burst. That would bring those buried French linkages back to the surface so they wouldn’t have to compete with my Spanish.

I tested that hypothesis by taking a week to visit France and speak only in French (luckily only an inexpensive train ride away from Spain). And, despite the initially bumpy start, things became easier until by the end of the week until it was only a little worse than my prior ability. I made a few more mistakes and had some tip-of-the-tongue moments, but otherwise I regained my ability after only a few days.

Upon coming back to Spain, I found myself tripping a little bit in Spanish, but to a far lesser extent than I had noticed in French.

Observations and Switching Costs

Beyond my personal experience, I’ve also made some informal observations about how this problem seems to affect other language learners.

First, it seems to affect you less when your language ability is higher. My French was an upper intermediate level when I last left France, but not at the point of mastery. Learners I’ve met who speak at nearly bilingual levels don’t seem to suffer the overwrite problem as badly. I think this is probably because the correct words are imprinted so automatically that they are very difficult to bury beneath a new language.

Second it seems to affect people engaging in complete immersion more. Because I spoke almost entirely in Spanish for the better part of three months, I wasn’t learning how to switch between French and Spanish. Had I learned in an environment where I still needed to use French frequently, I would have learned that skill as my Spanish improved.

Thirdly, it seems more of an issue the closer the languages are. People who have learned to very different languages seem to suffer from interference less than two similar ones. This is particularly relevant to me because Portuguese, my upcoming language, is very similar to Spanish.

Finally, switching costs seem to diminish the more they are practiced. Believe it or not, my French actually suffered the most at the beginning of my trip in Spain, despite Spanish only having a small amount of time to form an impression. Even the small amount of practice in French I did seemed to make switching easier.

Potential Solutions to Language Switching Costs

I can’t say that I have this problem solved yet, but I do have some strategies I’m hoping to use in the next three countries.

Benny Lewis suggests some solutions here that are worth exploring: namely practicing switching and using your body language to reinforce the distinctions between languages.

My feeling is that part of the problem I faced with French was that I was learning Spanish, but I wasn’t learning the necessary skills to keep it separated from French. For example, I had learned “il y a” => “there is” when I learned French, and “hay” => “there is” in Spanish, but I hadn’t learned NOT to use “hay” => “there is” in my French.

This suggests that the best way to avoid switching costs is to practice switching. In some senses, this means “re-learning” a bit of your previous language in order to keep it separate from your current one.

My Plan for Portuguese, Chinese and Korean

Given I have three more languages to learn, and I certainly don’t want to end the trip with each language having replaced the one preceding it, I think practicing switching will be essential to maintain all of the languages.

Since switching costs seem higher in more similar languages, I think this means Brazil will be the country I’ll need to put the most emphasis on. My plan is to do an hour or two of tutoring every week in both French and Spanish, in addition to my Portuguese. That should give me practice switching while I learn the new language.

With Chinese and Korean this might not be possible—the amount of work to reach an acceptable conversational level might be too high. If so, this will mean switching costs will be something I’ll have to practice on when I return to Vancouver.

Fortunately, more similar languages are easier to learn, but harder to keep separated. That means that if Brazilian Portuguese is significantly easier than Chinese to learn, I’ll have more time to focus on maintaining that separation. Conversely, if Chinese is harder to learn, it might also be harder to mix up.

Does This Mean You Shouldn’t Learn More than One Language?

The decision to learn one language extremely well or a couple languages to conversational abilities depends on a lot of different factors. If you’re interested in traveling and speaking with locals, I feel a rough conversational ability can get you most of the way. If you’re looking to live and work in a single language, then mastery is often worth the investment.

True mastery also dwarfs switching costs in terms of time investment. A decent conversational ability is perhaps 10% of the work of reaching a mastery level. Switching costs may make learning multiple languages harder, but it is still considerably less work than mastery. But it does mean that, if you plan to learn multiple languages, that you should have a plan in place to keep them separate.

What are your thoughts on the switching problem? If you’ve learned more than one language before, did you experience moments where you got them confused? What was your solution? Share your thoughts in the comments.