Down with Forced Speech

Stephen Krashen
University of Southern California (Emeritus)
skrashen@yahoo.com

When acquirers are forced to produce language that they have not yet acquired, known as
“forced speech,” they often experience anxiety. I argue here that forced speech is not only
uncomfortable, it makes no direct contribution to language acquisition.

Keywords: Forced speech, comprehensible input, anxiety.

INTRODUCTION: ANXIETY AND FORCED OUTPUT1

When my daughter was about five years old, she would occasionally play with a neighbor’s
child of about the same age, and the parents would take turns being with the children. On one
occasion, I went to the neighbor’s house to pick up both girls, while our neighbor went off to the
local community college to attend a Spanish class. Just before she left, she dashed into the
kitchen and poured herself some water and took a pill, clearly in a hurry. I asked her what the
hurry was. She told me: “I just took a valium. I had to. It’s Spanish class, it freaks me out.”
Ever the researcher, I asked her what it was about Spanish class that made her so nervous. She
told me if was having to speak in class – being called on or doing an oral report.
This reaction to forced speech agrees with reports from the research: Price (1991) interviewed
a group of ten foreign language students in US who considered themselves to be anxious about
foreign language study. When asked what bothered them the most about foreign language
classes, students said that their greatest source of anxiety was having to speak the foreign
language in class.” Loughrin-Sacco (1992) reported that for every student in a beginning French
class, “speaking was the highest anxiety-provoking activity. Four of the five activities (out of 21)
rated as the most anxiety-provoking by Young’s subjects (Young, 1990) entailed speaking.
I posted the anecdote about our neighbor and her reaction to Spanish class on facebook on
August 11, 2018, under “Story Listening and Reading for Language Acquisition.” The reactions
added valuable depth.
The posts all confirmed that forced output produces high anxiety in students in foreign
language classes.
Some of these reports described experiences that took place about the time the above research
was published.
Judith Dubois: When I was teaching English Composition in the university in Agen
(France), I had classes of mostly girls who wanted to become English teachers. One day … I
asked them to write about a frightening experience. I was surprised that almost a quarter of the
class wrote about being called on to speak in class as the most frightening thing that had ever
happened to them.
Kathrin Shechtman: Forced output is terrible. It makes students feel dumb, especially since
they will be corrected. School almost ruined English for me, and it made me hate French. It made
me feel stupid. I don’t want to do this to my students.
Others confirmed that the problem is still with us, in today’s classrooms:
 Kyung Sook Cho: “I see that the majority of students in my (university) class (in Korea)
have developed a great deal of anxiety over English class. Everybody has it. They feel relieved
when they share their feelings about this, and realize that they are not the only ones.”
Jen Schongalla noted that some administrators and scholars believe in and insist on forced
output. “I would love to read these (facebook) posts to my administrators and to other random
‘experts’ who still say things like ‘make them talk’ or ‘get them speaking’ etc. “ Schongalla
adds “I have a hard time communicating this concept despite explaining theory, pointing to
diagrams, etc. The problem of forced output is the one thing everyone seems to willfully ignore,”
which leads to the question of whether people disagree with the research, and/or have never
experienced the discomfort of forced output.
Beniko Mason made an important distinction between private personal conversation
and public displays of competence: “When my students come up to me after a Story
Listening lesson and want to talk with me in English about the story, that is wonderful.
Sometimes they write me a note or a message telling me what they personally think about
the story. I do not have my students talk in my class, as my students are still beginners
(TOEIC 250 to 600). They are shy about speaking in English. They understand a lot, but
they are still shy about talking.” Note that traditional methods demand public displays of
speaking immediately.

DOES SPEAKING CONTRIBUTE TO LANGUAGE ACQUISITION AND/OR SPEAKING

FLUENCY?
Correlational studies show no clear advantage to encouraging speaking
A small number of studies show positive correlations between second language
proficiency and student oral output in classes as well as outside of school (reviewed in
Chaudren, 1988). As Chaudren points out, however, since these studies are correlational,
this relationship could be the result of increased competence, not its cause. Other studies
show no relationship between amount of output and competence. Ely (1986) reported that
voluntary class participation in a university Spanish class correlated with performance on a
test of oral fluency (r = .40) for first quarter students but not for second quarter students (r
= -.02) and correlations between voluntary participation and performance on measures of
oral and written correctness were small and insignificant. Ross (1992), in a study of adult
EFL in Japan, found no relationship between the amount of pair work practice and fluency.

Delaying speech production

Gary (1975) provides evidence showing that delaying speech production has no negative
effect in classroom instruction: She examined children studying Spanish as a foreign language
over a period of five months. Her experimental group did not speak at all for the first 14 weeks
but, instead, had to produce "active responses" that demonstrated comprehension. Also, they
were not forced to speak for much of the next seven weeks. The experimental group was shown
to be superior to the control group in listening comprehension and equal in speaking, despite the
fact that the controls had more "practice" in speaking.

How to promote speaking fluency

Theory says that competence is clearly related to input, not to output. It predicts that
speaking ability (fluency and accuracy) is the result of obtaining comprehensible input. Sari
(2011) provides evidence for this, showing the effect of written input (reading) on
speaking. Sari compared two classes of intensive English of high-school students in Turkey.
Neither class placed an emphasis on speech production, but the class that did more self-
selected reading and listening (and less grammar study and less writing) did significantly
better on a test of speaking (picture description).
The enduring demand for premature speaking appears to be correlated with a
reluctance to embrace or even consider an obvious and highly effective source of
comprehensible input: pleasure reading (Cho and Krashen, in press).

Informal environments: acquisition proceeds quite well when speech is not forced.
Successful language in the informal environment, outside of class, often includes a silent
period, or period of very reduced production. Acquirers are not forced to speak and may
remain silent for month or years, but there is evidence that they do well in language
acquisition, consistent with the view that production of language does not require output.
Typical reports have been provided by Hakuta (1974) and Ervin-Tripp, 1974),
Here are some less typical reports.
The indigenous people living in the Vaupes River area in Columbia and Brazil, when they
were studied by Sorenson in the 1960’s, spoke about 24 languages, tremendous diversity for a
group of only 10,000 people. Multilingualism is stimulated by an unusual custom: People are
required to marry someone who does not speak the same language! Sorenson points out that
although some of the languages spoken in this area were mutually intelligible, requiring only a
few days listening to understand, others are not, even some of those that are closely genetically
related (Sorenson, 1967. p. 674).
In this area, children grow up with the father’s and mother’s languages, but during
adolescence, according to Sorenson, an individual “actively and almost suddenly” begins to
speak the two or three other languages he or she has been exposed to. “In adulthood he may
acquire more languages; as he approaches old age ... he will go on to perfect his knowledge of all
the languages at his disposal” (p. 678).
Most interesting for this discussion is how these multilinguals go about the task of language
acquisition. According to Sorenson, “The Indians do not practice speaking a language they do
not know well yet. Instead, they passively learn lists of words, forms, phrases in it and
familiarize themselves with the sound of its pronunciation... They may make an occasional
attempt to speak a new language in an appropriate situation, but if it does not come easily, they
will not try to force it.” (p. 679-80). Sorensen was told that “it takes from at least one to two
years to learn a new language fluently” (p. 680). In other words, speech is not forced.
Nida (1957) tells us “I have personally inquired of a number of African polyglots just how
they learned the language of neighboring tribes. Almost without exception, the story is the same:
they went to live in a neighboring village, or on some plantation, or in the mines they were
working with people who spoke another language. But instead of trying hard to learn the
language, they seemed to just take it for granted that after listening to the language long enough,
they would find that they could ‘hear’ it. ‘We just live there and listen, and before we know it,
we can hear what they say. Then we can talk,’ one African explained. This does not mean that he
expected to be able to understand (i.e. ‘hear’) everything in the language before he said anything,
but his whole attitude was one a passive absorption, confident that his ears and brain would take
in the language and that, without particular worry or concern on his part, he would be able to
understand and to speak sooner than even he imagined. (see Krashen 2017, for a similar
assumption made by polyglot Steve Kaufman).
 There are obvious similarities between these cases and that of a young immigrant, Armando,
who arrived in the US from Mexico in his teens. I interviewed Armando after he had been in the
US working at an Israeli restaurant for 12 years (Krashen, 2016). Armando spoke English well,
but spoke Hebrew as well or better. Armando told me that it was two or three years until he was
comfortable in conversation even though he heard Hebrew all day on the job. He said that he
never forced or pushed himself with Hebrew, that his approach was relaxed. This was made
possible because of his friendly relationship with members of the family that owned the
restaurant and customers. He had never studied Hebrew grammar, had no idea what the
grammatical rules were, and only received correction on vocabulary. Armando could not read
Hebrew.
With his permission, I recorded a short conversation in Hebrew between Armando and a
member of the family who owned the restaurant and played it for four educated native speakers
of Hebrew not associated with the restaurant. The evaluations were very positive: The range was
from “very good” (not a native speaker but clearly fluent and comfortable in Hebrew) to
“native.”
Clearly, language acquirers can reach high levels of competence without being forced to
speak.

Allowing speech to emerge in classrooms.

Teachers attest that in foreign language classes in which comprehensible input is provided,
speaking ability does not manifest itself immediately, but “emerges” gradually. There appears to
be individual variation in the length of the silent period, but there is no evidence this has an
impact on improvement in the second language.
Varvel (1979) describes a silent period in formal instruction (Silent Way methodology) that
lasted for nine weeks, indicating that there may be a fair amount of individual variation in the
duration of the silent period for adults in language classes: "There was a woman from Taiwan
who after several weeks was still conspicuously silent in class. She never talked, and when called
upon would only answer in a whisper, saying only what was required. It was clear, however, that
she was one of the most attentive students in the class, had a clear understanding of what was
being done, and seemingly enjoyed the class. She also had a positive attitude towards what and
how she was learning. At no time was she coerced into active participation.
"Then one day in the ninth week of school she sat in the front row and actively participated
throughout the whole hour. From that point on, she continued to participate actively in a more
limited way and at times helped others and was helped by others..." (p. 491)
While there may have been other reasons for this student's silence, this example suggests that
the silent period should be respected, and that some students develop speaking readiness later
than others.

SOME CONCLUSIONS AND IMPLICATIONS

The analysis presented here supports the Pleasure Hypothesis (Krashen, 1994):
Activities that are perceived to be unpleasant do not help language acquisition, and
activities that promote language acquisition are perceived to be pleasant. (But the Pleasure
Hypothesis does not claim that all activities that are pleasant are beneficial.) Specifically,
forced output causes anxiety, and does not help language acquisition. Lao and Krashen
(2008) confirmed this hypothesis, presenting data showing that activities that middle
school students of Mandarin as a foreign language rate as more effective were also
considered to be more pleasant (r = .78). Consistent with language acquisition theory,
effective and pleasant activities were those that supplied interesting and comprehensible
input.
One practical implication is obvious: Don’t require speaking. But we might also consider
another policy change: Don’t test speaking.
If we test speaking, the result will be to encourage more speaking: But as argued here, more
speaking does not result in more language acquisition. Rather, according to the Comprehension
Hypothesis, the ability to produce output is the result of getting comprehensible input. In
addition, testing speaking is enormously time-consuming (and expensive) to administer and
score.

DEFENDING OUTPUT
Domain for Error Correction
It can be claimed that output is valuable because it provides a domain for error correction:
When we speak, we make mistakes, which others correct, which helps our acquisition, helping us
arrive at a more accurate form of the rule. This is, of course, conscious learning, not language
acquisition. But a number of studies show that correction produces only a fragile effect that lasts
for only a short time (e.g., Truscott, 1999; see also Krashen, 1994a, for a review of studies
showing that correction has little effect). Also correction is not frequent in classrooms or in the
informal environment (Krashen, 1994a, pp. 60-61).

Domain for Comprehensible Output

The comprehensible output (CO) hypothesis states that we acquire language when we attempt
to transmit a message but fail and have to try again. Eventually, we arrive at the correct form of
our utterance, our conversational partner finally understands, and we acquire the new form we
have produced. In Krashen (1998) I argue that:
1. Instances of comprehensible output, where the native speaker requested confirmation or
clarification, are quite rare.
2. High levels of linguistic competence are possible without output.
3. There is no direct evidence that comprehensible output leads to language acquisition.

Speaking invites comprehensible input.

When we speak, somebody might answer (conversation), which could result in more
comprehensible input. It has yet to be investigated how much comprehensible input we receive
in conversation, and to what extent this input is helpful. Most important, as noted above,
acquisition can take place without interaction, from reading and listening alone.

A FINAL ANECDOTE
Sometimes, as part of my public presentations, I include some sample lessons in a
language that is (usually) unfamiliar to the audience. Before I start, I ask the audience if it
is ok with them if I give them the lessons, and at the same time I come out from behind the
speaker’s podium and walk a few paces into the area where the audience is sitting. The
reaction is the same, and has been the same since I starting the demonstration about 30
years ago: Complete silence, and no movement. No looks of anticipation, no nodding of
heads. I then comment, “What enthusiasm.” Then I ask: “What goes through your mind
when you are listening to some ‘expert’ give a talk, and suddenly he announces that he will
give you a language lesson, comes out from behind the podium and comes right over to
where you are sitting? And then they tell me: “Fear” or “panic,” or “you are going to call on
me.”
I point out that they are professional educators, and yet the idea of a language lesson in
public makes them nervous. I ask them “What does this mean?” And then I answer for
them: It means that we are doing something profoundly wrong and unnatural in language
teaching. I tell them that anxiety in this situation is not their fault. I would feel the same
way. Then I announce that I will give them the lessons anyway, but I that I will return to the
front of the room. There are obvious signs of relief.

NOTES
1. I assume in this paper some familiarity with the Comprehension Hypothesis and
other aspects of language acquisition theory (e.g. Krashen, 1981, 1982, 2003;
Krashen, Lee, and Lao, 2017. )

REFERENCES
Chaudren, C. (1988) Second language classrooms: Research on teaching and learning.
Cambridge: Cambridge University Press.
Cho, K.S. & Krashen, S. The power and pleasure of reading. Lanugage Teaching and Learning
(in press).
Ely C. (1986). An analysis of discomfort risk-taking, sociability, and motivation in the L2
classroom. Language Learning, 36, 1-25.
Ervin-Tripp, S. (1974). Is second language acquisition like the first? TESOL Quarterly, 8, 111-
127.
Gary, J.O. (1975). Delayed oral practice in initial stages of second language learning. In M. Burt
and H. Dulay (Eds.) On TESOL ’75: New Directions in Second Language Learning and
Bilingual Education. Washington TESOL. pp. 89-95.
Hakuta, K. (1974). A preliminary report of the development of grammatical morphemes in a
Japanese girl learning English as a second language. Working Papers on Bilingualism, 3,
18-43.
Krashen, S. (1981). Second language acquisition and second language learning. New York:
Prentice-Hall. (free download at sdkrashen.com)
Krashen, S. (1982). Principles and practice in second language acquisition. New York:
Pergamon Press. (free download at sdkrashen.com)
Krashen, S. (1994a). The input hypothesis and its rivals. In N. Ellis (Ed.) Implicit and explicit
learning of languages (pp. 45-77). London: Academic Press.
Krashen, S. (1994b). The pleasure hypothesis. In J. Alatis (Ed.) Georgetown University Round
Table on Languages and Linguistics (pp. 299-322). Washington, DC: Georgetown
University Press.
Krashen, S. (1998). Comprehensible output? System, 26,175-182.Krashen, S. 2003. Explorations
in language acquisition and use. The Taipei Lectures. Portsmouth, NH: Heinemann.
Krashen, S. (2017). Polyglots and the comprehension hypothesis. Turkish Online Journal of
English Language Teaching (TOJELT), 2(3), 113-119.
Krashen, S. Lee, SY, & Lao, C. (2017). Comprehensible and compelling: The causes and effects
of free voluntary reading. Santa Barbara: Libraries Unlimited. ABC-CLIO, LLC.
Lao, C. and Krashen, S. (2008). Do students like what is good for them? An investigation of the
pleasure hypothesis with middle school students of Mandarin. International Journal of
Foreign Language Teaching, 4(2), 19-20.
Nida, E. (1957). Learning a foreign language. New York: Friendship Press.
Price, M. L. (1991). The subjective experience of foreign language learning anxiety: Interviews
with anxious students. In E. K. Horwitz & D. J. Young (Eds.), Language anxiety: From
theory and research to classroom implications (pp. 166-172). Englewood Cliffs, NJ:
Prentice Hall.
Ross, S. (1992). Program-defining evaluation in a decade of eclecticism. In J.C. Alderson & A.
Barettta (Eds.) Evaluating second language education (pp. 167-195). Cambridge:
Cambridge University Press.
Sari, R. (2013). Is it possible to improve writing without writing practice? International Review
of Foreign Language Teaching 8, 6-10.
Sorenson, A. (1967). Multilingualism in the northwest Amazon. American Anthropologist, 69(6),
670-684.
Truscott, J. (1999). What’s wrong with oral grammar correction. Canadian Modern Language
Review, 55, 437–456.
Varvel, T. (1979). The Silent Way: Panecea or pipedream? TESOL Quarterly,13, 483-494.
Young, D. (1990). An investigation of students’ perspectives on anxiety and speaking. Foreign
Language Annals 23, 539-553.