Just a quick follow up. I passed the question on to the MultiEd-L list
and had a very helpful reply from Jill Kerper Mora, which I copy
below. First I would mention that there was a passage in a UNESCO
report that I posted on this list a while back in message #79 that
gives an ambiguous impression of what they mean by "subtractive." I
read it to mean bilingual ed. that is used mainly or only to "move
students on" to literacy in an LWC, with no plan for
bilingual/multilingual proficiency or literacy.
"In regions where the language of the learner is not the official or
national language of the country, bilingual and multilingual education
can make mother tongue instruction possible while providing at the
same time the acquisition of languages used in larger areas of the
country and the world. This additive approach to bilingualism is
different from the so called subtractive bilingualism which aims to
move children on to a second language as a language of instruction."
(UNESCO, "Education in a multilingual world, "
, pp. 17-18)
Another related issue is whether the late-exit strategy of L1 (or
other dominant local language) instruction, as opposed to early exit,
makes it less "subtractive." At this point I'm not arguing either way,
but do want to suggest that the use of "subtractive" - a term with
obvious negative connotations - is contested and and in any event may
be misleading to use to describe the old monolingual French-only (or
English-only) models. I used "immersion" in my earlier comments to
describe the latter, but as Dr. Mora points out, that use is also
contested. (Some have, only half-jokingly, suggested "submersion" for
monolingual L2 instruction models.)
From Jill Kerper Mora on MultiEd-L :
"What determines whether or not a program is considered to be based on
an additive or subtractive approach is determined by its goals and
objectives in regard to bilingualism. Subtractive programs see L2
learning as the major and ultimate goal of instruction. These programs
are not concerned with building and or/preserving L1 abilities and
skills because of their inate value to the learner, but may employ L1
as a tool for learning L2. Therefore, transitional bilingual education
programs are often considered subtractive because L1 is used
temporarily as a tool for learning L2 and eventually phased out.
Bilingualism and/or biliteracy per se are not the objective of such
programs, even though they may be an artifact of bilingual
instruction. The two program models considered to be additive are
maintenance (also called developmental bilingual programs) and dual
immersion (two-way bilingual education). In actuality, the term
"immersion" was used by second-language and foreign language educators
to describe additive programs that serve a homogeneous group of L1
speakers of a particular language or mixed groups with two different
L1s taught by fully proficient bilingual teachers with the goal of
adding proficiency in the L2 while maintaining skills in L1. The term
has been so successfully co-opted by proponents of English-only
instruction that it has limited use now among bilingual educators. So,
yes, this term can and usually does refer to subtractive programs.
"I hope this helps. See my web page on this topic at this URL:
--- In Multilingual_Literacy@yahoogroups.com
, "Donald Z. Osborn"
> Thank you Barbara and 'Tope for your comments. It seems we have a
> issue or two here. 'Tope brings up some more complicated
> context that we should take up, but I wanted to address the quote
> kindly supplied.
> As a non-expert in this field,* I been given to understand that
there is a
> different definition of "subtractive" than the one below. What they call
> "subtractive model" I've heard called "immersion." What they call
> model(s)" I've understood to be "subtractive bilingual approach(es)"
> instruction is primarily a way of getting to L2 instruction, and L1 is
> abandoned as a language of instruction or discourse in favor of L2).
> approach in the latter schema is "additive bilingual," in which L1
> abandoned but continues to be used along with L2.
... [rest deleted]