Multilingualism & Disorders: How Can We Know When It Is Time to Call the Speech-Language Therapist?

by expert · 7 comments

Multilingualism is neither odd nor worrying: what is odd, and very worrying, are the views that persist about it.

The topics in this Multilingualism & Disorders series aim at clarifying the misconceptions that associate multilingualism with disorders.

Each topic offers a brief introduction to common questions, and includes one token reference, which either marks watershed findings or otherwise addresses points which are perhaps less known within research on multilingualism.

This condensed format is deliberate, meant to invite discussion, thoughts, and more queries. Only with your help, as engaged readers, can we make this series the useful tool that we hope it will become.

— Madalena Cruz-Ferreira

How can we know the difference between normal multilingual language development and when we should see a speech-language therapist?

There are so far no norms devised for multilingual development. Extant norms are based on monolingual development, that is, on typical development in different single languages. Since multilingualism means differential use of different languages, monolingual norms can only be of limited use in the diagnosis of speech-language disorders among multilinguals.

As parents, we can check for signs of language disorder in the same way that we check for other signs of disorder in our children.

We do this by comparing the child to itself. A child who is less lively than usual may be running a fever; a child who is using fewer words than usual or has settled down to use just a reduced number of words altogether may be having language problems.

A language disorder affects the whole of the child’s linguistic repertoire, whether the child speaks one language or more than one. For multilingual children, this means that language disorder will affect all of the child’s languages.

Having fewer words in one language rather than another, or preferring to use one language rather than another in specific situations or with specific people is typical of multilingualism and is therefore not a sign of linguistic disorder.

Specific reference for the above topic:

General references relevant to core issues in the Multilingualism & Disorders series.

General public:

  • Cruz-Ferreira, M. (2010). Multilinguals are …? Battlebridge Publications. Book URL:
  • Grosjean, F. (2010). Bilingual. Life and reality. Harvard University Press. Book URL:


  • Cruz-Ferreira, M. (Ed.). (2010). Multilingual norms. Peter Lang.
    Book URL:
  • Genesee, F., Paradis, J., & Crago, M. (2004). Dual language development and disorders: A handbook on bilingualism and second language learning. Brookes Publishing.
    Book URL:
  • Grosjean, F. (2008). Studying bilinguals. Oxford University Press.
    Book URL:
Madalena Cruz-Ferreira, PhD, University of Manchester, UK, is a multilingual parent, educator and scholar, and the author of Multilinguals are...?, a book on myths and misconceptions about multilingualism. Her blog Being Multilingual deals with multilingualism at home, in school and in clinic. Her contact, and details on her work, are at:

Disclaimer: This post and the comments provided below have been provided for informational and entertainment purposes only and is not intended to replace or substitute for any professional financial, medical, legal, or other advice. This post has been published with the full consent of the author. The author has agreed to Madalena Cruz-Ferreira answering the Ask Madalena question publicly as well as readers leaving comments in the comment section below. Multilingual Living makes no representations or warranties and expressly disclaims any and all liability concerning any treatment or action by any person following the information offered or provided within or through this and any other information on this website. If you have specific concerns or a situation in which you require professional or medical advice, you should consult with an appropriately trained and qualified specialist. Please read our Terms of Use for more detail or contact us with any questions.

{ 7 comments… read them below or add one }

1 Jefferson September 27, 2010 at 8:14 am


first of all, thanks for your website and for this article.

Second, I think the links to the Stow & Dodd’s paper don’t work (didn’t for me); I’d suggest, for the abstract and pdf, respectively:

And, finally, when you say we should compare the child with herself, are you really suggesting parents should not care at all to milestones? After all, normality is, by definition, a matter of comparison with averages, with ranges, i.e., with other kids.

Best regards,
Jefferson Portela


2 Corey September 28, 2010 at 7:09 am

Thank you for pointing out the out of date links, Jefferson! Very appreciated! I have updated them based on Madalena’s links listed in her comment below. Thank you for the additional links in your comment!


3 Alice September 27, 2010 at 8:48 am

Hi Jefferson, you write “After all, normality is, by definition, a matter of comparison with averages, with ranges, i.e., with other kids.”

I think one should add the word “monolingual” before “kids” – linguistic “normality” is usually defined according to the linguistic behavior of other MONOLINGUAL kids. Thus also probably those milestones-charts, which are usually based on monolingual speech development. Someone correct me if I’m wrong, though.

From what I understand, there also seems to be a big window around those milestones that define “normality”….


4 Jefferson September 27, 2010 at 11:27 am

Hello Alice,

I’m just trying to gather information on the subject and was curious about your opinion.
Yes, from what I’ve seen so far, these ranges are quite wide – sensibly.

But I am not so sure about those charts being restricted to monolingual children. For example, it would make sense that nonverbal milestones are the same; and when they say something like “By age 2, most children can use two-word phrases”, why shouldn’t that apply also for bilinguals? Also, when a vocabulary of a certain size is mentioned, couldn’t it be enough to combine the vocabulary of all the languages the kid speaks (in order to compare)?
I found some who support those views:
– language developmental milestones are the same:
– it makes sesnse to consider languages combined:

Yet, I am not completely convinced. These views sound like they could be too simplistic, too linear – as they seem to ignore interactions between the different languages the child is exposed to. So I am still open and looking for information.


5 Alice September 28, 2010 at 12:42 am

Hi Jeffrey, GREAT questions and comments! Thank you for the links.

First off, I’m actually not the expert here (Madalena is, and I would love it if she could jump into the discussion as well) – but I have often wondered about this issue myself! I have often asked how one is to define a “norm,” and how one ought to define “normal linguistic behavior” for my multilingual child. What is normal??? A tricky question! I wasn’t disagreeing with your comment, but simply adding how I always understood the definition of a norm – usually with reference to a monolingual.

Regarding those milestone charts, again, I am not the expert, but from what I have read is that when it comes to EARLY milestones, like babbling, first words, two word speech, there does indeed seem no difference between bilingual and multilingual behavior. So, in agreement, it seems as though those early milestones may indeed be applicable to all children. (see Barbara Zurer, _Raising a Bilingual Child_, chapter 7:”Research comparing Bilinguals to Monolinguals”. A fantastic book, and a fantastic chapter! I recommend it highly!).

However, it gets tricky with later milestones and word count, specifically when those charts say, I don’t know, by age of so and so your child should produce 36.5 words and 5 sentences or whatever – and this is what I was referring to. Unless, as you rightly point out, one counts the sum of vocabulary in all languages, these charts can be fairly misleading with regards to a multilingual child. But even just counting “the sum of all vocab production in all languages” isn’t as straightforward as it sounds. Zurer writes that in a study that attempted to track bilingual children’s vocab growth in order to compare it to that of monolinguals, they ran into difficulties with regards to how one exactly is to measure this. Does one count word pairings? word forms? sounds? meanings? referents? only expressive vocabulary? or receptive vocabulary too? And they’d get different results of course if they counted Total conceptual vocabulary versus receptive vocab.

So here’s our problem: how do we, as parents, then go about counting these words as well? And what are we supposed to compare them to?

And what is one to make of this huge windows around the averages? Zurer writes 5 months on either side for first words, and six months on either side plus or minus for first phrases and two-word combinations. That is such a huge window that I can’t make anything of this measurement.

(On a personal note, I used to agonize over milestone charts and was considerably happier when I stopped using them as a reference).

Baker has an interesting milestone chart specifically for bilingual children in his book. I do like this one! 😀

0-1 year babbline, cooing, laughing (dada, mama, gaga)
Around 1 year: first understandable words
1-2 years: two word combinations, moving slowly to three and four word combinations. Three-element sentences (e.g. “Daddy come now”; “That my book”; “Teddy gone bye-byes”)
3 to 4 years: simple but increasingly longer sentences. Grammar and sentence structuring starts to develop. Conversations show turn taking.
4-year onwards: Increasingly complex sentences, structure and ordered conversation. Use of pronouns and auxiliary verbs.

from p. 18, Colin Baker, _Parents’ and Teachers’ Guide to Bilingualism_



6 Jefferson September 28, 2010 at 2:27 am

Hey Alice,

I see you have found out more on the subject than I have so far.
Thanks for sharing, especially for the Baker’s milestones chart . It does look very similar to monolingual ones, minus the word count – what is probably very good. 🙂 I guess I would not find it an easy job to count words even for a monolingual child, let alone a multilingual one… Well, that different charts often assign different number of words for the same age range should already make one suspicious indeed about these numbers.

About your question on ‘normality’, it’s indeed a tricky question. But, although we could philosophy a lot about it, I guess that, at least for this subject, we have to content ourselves with mere statistics: for a given population you compute the average and then call ‘normal’ what is, e.g., within two standard deviations of it – in other words, what applies to about 95% of the kids.
That reflects what we intuitively do: if most one-year-olds we’d heard about could communicate better than of our two-year-old, we’d be worried, and probably rightly so. Of course, the question of whether the statistics are well-done enough is a completely different one – and certainly one always to be kept in mind.


7 Madalena Cruz-Ferreira September 28, 2010 at 6:28 am

Hello Jefferson and Alice, and many thanks for your comments.
First of all, I must apologise for the broken URLs. I checked them out latest in August and they were working fine. Here are two currently working links:
Journal URL:
If all else fails, you can look up the article by its DOI (digital object identifier) at the DOI finder site,, by typing the article’s DOI in the search box: 10.1080/1368282031000156888

Now to the discussion. Alice, you are absolutely right in saying that the milestones that are available to us are based on monolingual development, and Jefferson, you are absolutely right in not being fully convinced by simplistic extrapolations of monolingual behaviour to multilingual behaviour. We would encounter exactly the same problems if we attempted to do the converse and measure monolingual development using multilingual milestones – which we cannot do because there are to date no normed multilingual developmental milestones.
The issue relates to Western(-like) traditions which assume monolingualism as “normal” and which go back all the way to the Ancient Greeks, who gave us the word ‘barbarian’ to designate anyone whose speech (= “babble”) wasn’t intelligible to educated monolinguals. The assumption lacks foundation in that multilingualism has both historical and statistical credentials: multilingualism can be traced through our earliest documental sources, and multilinguals outnumber monolinguals. In other words, there must be multilingual norms of linguistic, developmental and cultural behaviour, that is, there must be typical multilingual behaviour. The issue is then that we haven’t known how to look for these norms precisely because of monolingual assumptions.
Counting words, for example, is not straightforward even for specialists: see the provisos raised by Barbara Zurer Pearson’s research and mentioned in Alice’s latest post. Milestone charts for multi-word child productions, in turn, must allow for multi-language words, regardless of language. The issue here is that research, and so advice to parents, clinicians and educators, has been concerned with particular languages, that is, with each particular language of a multilingual, again on the assumption that particular single languages are what matters. To my mind, this makes little sense, because multilingualism is not about what happens to single, independent languages, it is about what happens when people have several languages available to them. As Jefferson wrote, “These views […] seem to ignore interactions between the different languages the child is exposed to”. Interaction is one core issue and, again, regardless of particular languages. Just like a monolingual is a monolingual regardless of particular language, so a multilingual is a multilingual regardless of particular languages.
These thoughts are the reason for my contention that the first step in deciding whether something may be amiss with our multilingual children is to compare the child to itself: we cannot fully rely on linguistic norms that do not reflect multilingual behaviour.


Leave a Comment

Previous post:

Next post: