J:N /-I; o« EL.- [;".

7 ~ 005)
j
) ;.j,q-N8o-o~ ~F IZ~S&71"Uli :HV
5.€c...oAJD ~6U-#c ~~C.H/;Vb AYVC t....e;-~tU/;ur;.
)'flAI-/v.J/4/"1 J1J J": ;C1"l L d~M..

31

Reading in a Second Language

David E. Eskey!
University of Southern California

INTRODUCTION

Reading and Second Language Acquisition

For second language learners, reading may be both a means to the end of acquiring the
language, as a major source of comprehensible input, and an end in itself, as the skill
that many serious learners most need to employ. Many students of English as a Foreign
Language (EFU, for example, rarely speak the language in their day-to-day lives but
may need to read it in order to access the wealth of information recorded exclusively in
the language. In complementary fashion, this reading can serve as an excellent source
of the authentic language students need to interact with in quantity-language that is
always meaningful, often in fully grammatical form, and that includes every feature
of the target language but pronunciation. Krashen (1993) claims that students who
read frequently

acquire, involuntarily and without conscious effort, nearly all of the so-called "language
skills" many people are so concerned about. They will become adequate readers, acquire
a large vocabulary, develop the ability to understand and use complex grammatical con-
structions, develop a good writing style, and become good (but not necessarily perfect)
spellers. (p. 84)

In recognizing the truth of this claim (what Krashen sums up as "the power of read-
ing"), the field has come a long way since the audiolingual era when reading and
writing were marginalized as "secondary reinforcement" for learning the spoken lan-
guage (Fries, 1945).

RESEARCH ON READING

L1 Research Applied to L2
Specialists in second language (L2) reading are often criticized for depending too
heavily on research in first language (Ll) reading, instead of focusing more narrowly

563
564 ESKEY

on studies of reading in a second language. There is some truth to this-especially ma.

in respect to language issues, which create far more problems for L2 readers than con
for L1 readers-but research on L1 reading provides a foundation for exploring both firn
the similarities and differences between L1 and L2 reading. Just as in language ac- the:
quisition research, where for years L2 specialists paid no attention to research in L1 for
language acquisition, despite its obvious relevance, similarities between these two this
kinds of reading far outweigh the differences. The differences are important and must ress
be addressed, but reading is reading in any context, just as language acquisition is hur
language acquisition. If human beings are the only living creatures who speak, they scht
are also the only living creatures who read, and they do so in much the same way and
throughout the world. Children acquire language instinctively (Pinker, 1994). They I
must be taught literacy, but all writing systems are language based and share some by!
fundamental characteristics. mo.
198·
An Overview of Research on Reading rear
thai
In the 1960s before the advent of what has come to be known as "cognitive psychol- pos
ogy," most research on reading was hardly worth-well, reading-because the then the
dominant behaviorist models in psychology could not accommodate discussion of Fro:
mental events, and reading is, almost purely, a mental event. One major exception not
was Edmund Burke Huey's (1908) The Psychology and Pedagogy of Reading, but this suo
sank without a trace in the behaviorist sea before being rediscovered in more recent pro.
times. Comparisons with linguistics (which Chomsky has always characterized as a and
branch of cognitive psychology) come to mind; but language behavior, in the form of thai
spoken and written discourse, has always been accessible to direct empirical investi- whi
gation, whereas reading produces almost no physical data for investigation, with the the
minor exception of eye movements. star
By the late 1970s,however, reading specialists like Kenneth Goodman (e.g, Good- J.
man, 1975) and linguists like Ronald Wardhaugh (1977) were disputing the notion reac
that reading is merely the passive side-complementing the active skill of writing- am
of being literate. Like the other receptive skill listening (with which it has much in as n
common as a psycholinguistic processing skill), reading is now generally understood up.
to be an active, purposeful, and creative mental process in which the reader engages thos
in the construction of meaning from a text, partly on the basis of new information pro- this
vided by that text but also partly on the basis of whatever relevant prior knowledge, Intr
feelings, and opinions that reader brings to the task of making sense of the words on "brr
the page. neu
Research in the 1970s and 1980s was characterized by a search for more accurate F
and more revealing models of the reading process. Major advances in our understand- of re
ing of what readers actually do when they read followed from a shift in orientation pra(
from commonsense bottom-up (from text to brain) models of the process-in which the and
reader is assumed to decode precisely (in the case of English) from left to right, from refe
letters into words, and from words into larger grammatical units in retrieving the wor
writer's meaning, step by step, from the text-to a totally different kind of model, the as C
so-called top-down (from brain to text) model of the reading process. This shift, some- like
times referred to as the top-down revolution-a movement in which Goodman and, to tl
later, Frank Smith (e.g., Smith, 1973)were especially prominent-generated massive poli
research support and was widely accepted, in one form or another, by reading spe- A
cialists everywhere. Top-down descriptions of the reading process characterize it as guis
what Goodman (1967)once called, in a famous remark, "a psycholinguistic guessing the]
game." The notion is that readers do not decode in precise or sequential fashion but Still
instead attack the text with expectations of meaning developed before and during the foeu
process, take in whole chunks of text (in short, jerky eye movements called saccades) Laru
 31. READING IN A SECONDLANGUAGE 565

-especially making use of just as much of the visual information on the page as they need to
.aders than confirm and extend their expectations-a process of predicting, sampling, and con-
loring both firming in which readers interact with texts by combining information they discover
nguage ac- there with the knowledge they bring to it in constructing a comprehensive meaning
earch in L1 for the text as coherent discourse. Since prior knowledge plays such a major role in
_these two this conception of reading, reading specialists also devoted considerable aHention to
.t and must research on schema theory, research that attempted to account for the way in which
juisition is human beings store and organize information in networks of related notions called
.peak, they schemata (or, more rarely, plans, scripts, or scenarios). In this area, the work of Rumelhart
same way and his associates was especially influential (e.g., Rumelhart, 1980).
994). They During the latter part of the 1980s, top-down models were increasingly challenged
hare some by proponents of interactive models of the process, who point out that strictly top-down
models cannot fully account for the results of much empirical research (Stanovich,
1980, provides an excellent summary)-research that, for example, shows that skillful
readers can process linguistic forms in print both more accurately and more rapidly
than less skillful readers can even in context-free situations where no prediction is
e psychol- possible, and that weaker readers are as likely as strong ones to guess at meaning on
e the then the basis of prior knowledge-both results that run counter to top-down assumptions.
.ussion of From such research results, advocates for these interactive models (which should
exception not be confused with the interactive reader /text process described earlier) infer that
;, but this successful reading entails a balanced interaction between bottom-up and top-down
ore recent processing skills, thus restoring the simple decoding of text to a more central role
rized as a and raising doubts about the guessing game metaphor. Although it might be argued
ie form of that these interactive models are simply modifications of the topdown approach,
3.1 investi- which do not involve the kind of radical new conception of the reading process that
,with the the movement from bottomup to top-down models entailed, they have become the
standard for discussions of the psycholinguistics of reading.
.g, Good- After this period of relative coherence during which most researchers focused on
ae notion reading as a psycho linguistic process, reading research in the 1990s splintered into
Nriting- a number of incommensurate perspectives. By the late 1980s, reading research had,
much in as noted, coalesced around interactive models, which give equal weight to bottom-
derstood up processing of texts (i.e., decoding) and top-down construction of meanings for
. engages those texts (i.e., comprehension). Stanovich (1992) provides an excellent example of
.tion pro- this balanced view. Today, however, the field is very much engaged in what Kamil,
owledge, Intrator, & Kim (2000) describe as both "broadening the definition of reading" and
vords on "broadening the reading research agenda" to include a wide variety of social, cultural,
neurobiological, and even political perspectives on reading.
accurate Recently, for example, many prominent researchers have moved beyond the study
.erstand- of reading as a psycholinguistic process to consider reading as a form of sociocultural
entation practice. These researchers are concerned with such questions as how much, what,
zhich the and why-as opposed to merely how-people read, if and when they do, with special
.ht, from reference to the reading behaviors of particular socioeconomic groups. Building on
ving the works on literacy by sociolinguists like Heath (1983) and Street (1984), such scholars
odel, the as Gee (2000) have begun to explore this sociocultural dimension of reading. Others,
:t, some- like Friere (Friere & Macedo, 1987) and Shannon (1996), have taken this perspective
tan and, to the level of "critical literacy" in which reading behavior is regarded as a form of
massive political behavior.
ing spe- At the same time researchers in cognitive science have moved beyond psycholin-
'ize it as guistic models toward work in neurobiology, and some studies have been done on
uessing the neurobiology of reading, especially in relation to dyslexia (Shaywitz et al., 2000).
lion but Still another direction is represented by studies of new technologies in reading that
ring the focus on the nature of reading in the rapidly expanding electronic media (Kamil &
accadesi Lane, 1998).
566 ESKEY

These many new perspectives on reading research have been accompanied by reading t
related changes in research methodology. Experimental research has largely given that this,
way to classroom-centered action research (McFarland & Stansell, 1993) and narrative motivatic
(Gilbert, 1993) and ethnographic approaches (e.g., Heath, 1983). There is less emphasis and was.
on the typical reader and more on targeted groups or individuals. Thus, protocol the subje
analyses (Pressley & Afflerbach, 1995) and case studies of single readers (Neuman & relation t
McCormick, 1995) have become increasingly popular.
In contrast to information-processing models, these newer perspectives have given It may
reading research a more human face but have also reduced the generalizability of its Kamil':
results and raised doubts about the internal coherence of a subject as broad and have to
complex as reading. knotolec
Since reading is a kind of experience (albeit in purely symbolic form), potentially reading
involving the entire range of any reader's thought processes, feelings, imagination,
Readii
and beliefs, as shaped by his or her genes and real-world experience, it is hardly
vocabula
surprising that no single model of reading behavior can dominate reading research
best (sorr
for long. Even the boundaries of the topic are fluid, reading being just one instantiation
for readii
of the larger concept of literacy. In the latest Handbook of Reading Research (Kamil et al.,
derstood
2000), essentially the bible of reading researchers, all seven of the book's subheadings
chicken (
are labeled as discussions ofliteracy, not reading. As the editors of that volume suggest,
counter r
the remarkable diversity of perspectives on reading characteristic of reading research
he calls"
today may either be described, optimistically, as "creating new frontiers of thought"
thought j
or, more skeptically, as "creating confusion" about a subject that probably cannot be
terial but
captured in a unified model.
and Coal
ers, addn
Research on Second Language Reading: Specific Issues a "necess
compreh
Whether in a first or a second language context, most reading research can be sub-
plex. Sue
sumed under three major headings, all of which have been mentioned in passing-
target lai
reading as a psycholinguistic process, reading as sociocultural practice, and reading
subseque
as individual behavior-because every human reader is, simultaneously, 0) a mem-
tended e:
ber of the species (a human reader who reads as humans do, as opposed to some
nition for
other kind of reader-a computer, for example), (2) a member of a network of socio-
this iSSUE
cultural groups (possibly, in relation to the writer, a member of a radically different
show tha
culture), and (3) an individual (and thus, within the limits established in (1) and (2),
process f,
cognitively and affectively unique to some extent). Within this general framework,
skills.
research in the following specific areas seems especially pertinent to major issues in
Many
the field of second language reading.
context a:
strategyi
Reading as a Psycho linguistic Process Similarly.
not, freqr
Reading and Language Proficiency. Not surprisingly, the variable that correlates
the best r
best with success in second language reading is proficiency in the language. Motiva-
readers iJ
tion and background knowledge (of content) are also important, but reading begins
to a kind
with decoding of language; and reading comprehension, although it involves both
reading 2
bottom-up and top-down processing, begins with, and so depends on, rapid and ac-
readers sJ
curate decoding of the text (Birch, 2002). In any discussion of second language reading,
or be tau]
the place to begin is thus proficiency in the language. Serious discussion of this issue
Vocabula
can be traced to Clarke's (1980) "short-circuit" hypothesis, which challenged the no-
read text,
tion that skillful readers in one language could simply transfer their skills to reading
in a second language (the so-called "language interdependence" hypothesis; see, e.g.,
Cummins, 1984). Clarke argued for a language proficiency "threshold" i readers whose Readit
knowledge of the target language fell below that threshold, no matter how proficient for succe:
in their first language reading, could not transfer their skills to their second language tangling J
 31. READINGIN A SECONDLANGUAGE 567

npanied by reading until they had mastered more of the language. Hudson (1982) demonstrated
.rgely given that this threshold cannot be identified in absolute terms but varies with the reader's
ad narrative motivation and knowledge. Clarke's basic premise however has stood the test of time
ssemphasis and was. addressed most extensively in Alderson's (2000) book-length treatment of
JS, protocol the subject. With reference to the competing hypotheses, Carrell (2001) observes, in
(Neuman& relation to a more recent study:

have given It may be more profitable to think of these two hypotheses in terms of Bernhardt and
ibility of its Kamil's restatements of them: "How first language (LI) literate does a second language reader
broad and have to be to make the second language knowledge work?" and "How much second language (L2)
knowledge does a second language reader have to have in order to make the first language (LI)
potentially reading knowledge work?" (Bernhardt & Kamil, 1995, P: 32)
nagination,
.t is hardly Reading and Vocabulary. The relationship between reading and knowledge of
19 research vocabulary is well-documented and reciprocal. It is now well understood that the
stantiation best (some would argue the only) way to acquire the extensive vocabulary required
.amil et al., for reading widely in a second language is reading itself, and it is equally well un-
bheadings derstood that a prerequisite for such reading is an extensive vocabulary-a classic
re suggest, chicken and egg situation. Edward Fry (1981) has claimed that readers who en-
19 research counter more than one unknown word in twenty in a text will be reading at what
f thought" he calls "frustration level" and will thus be unlikely to continue reading-a sobering
cannot be thought for teachers of adult second language readers who want to read adult ma-
terial but often lack the vocabulary required to do so successfully. Huckin, Haynes,
and Coady (1993) provide a good summary of research for second language read-
ers, addressing most of the major issues, starting with automatic word recognition,
a "necessary but not sufficient" condition (Stanovich, 1991) for successful reading
m be sub- comprehension. For second language readers, the issues are obviously more com-
passing- plex. Such readers are often slower and less automatic in recognizing words in the
dreading target language than first language readers are (Favreau & Segalowitz, 1983). In
l) a mem- subsequent studies, Segalowitz, Segalowitz, and Wood (1998) have shown that ex-
1 to some tended experience in reading a second language has positive effects on word recog-
: of socio- nition for adult subjects, and Geva, Wade-Woolley, and Shaney (1993) have explored
different this issue with younger learners learning to read in two languages. Both studies
) and (2), show that the development of fast and accurate word-recognition skills is a complex
mework, process for second language readers that involves a wide range of knowledge and
issues in skills.
Many texts for the teaching of second language reading promote guessing from
context as a major means of decoding unknown words, but research suggests that this
strategy is overrated and often leads to misidentifications (Bensoussan & Laufer, 1984).
Similarly, research demonstrates that second language readers, whether guessing or
orrelates not, frequently misidentify words (Bernhardt, 1991). Thus, although reading remains
Motiva- the best means of acquiring a larger vocabulary, care must be taken not to immerse
g begins readers in texts that are lexically beyond them, which does in fact reduce reading
ves both to a kind of guessing game. The reciprocal relationship mentioned earlier between
land ac- reading and vocabulary must, in practice, be handled with care: Second language
reading, readers should be lexically prepared for any texts assigned and the texts should meet,
his issue or be taught in such a way as to meet, Krashen's i + 1 standard for comprehensibility.
l the no- Vocabulary cannot be force-fed through reading, and second language readers cannot
reading read texts that are lexically beyond their proficiency.
,ee, e.g.,
swhose Reading and Grammar. Although a firm grasp of syntax is obviously required
'oficierit for successful decoding, researchers in this area have not been successful in disen-
nguage tangling knowledge of sentence structure from other kinds of knowledge-especially
568 ESKEY

knowledge of vocabulary-to determine its particular contribution to reading com- Re

prehension. Some work has been done on syntactic simplification and elaboration,
but the results are inconclusive. Carrell (2001) summarizes: SUo
the IE
... the research on syntactic simplification and elaboration shows that while syntactic ing cc
simplification can enhance foreign or second language reading comprehension, the pic- infon
ture is a complex one, with linguistic complexity interacting with such other factors as reade
age, proficiency level, cultural background knowledge, and, possibly, item type. sense
have:
Reading and Text Structure. At the level of discourse, research suggests that that n
knowledge of text structure contributes to reading comprehension (Carrell, 1992; Ri- mean
ley, 1993). That is, a reader who understands the way in which the kind of discourse collec
he or she is reading is typically organized and will find it easier to comprehend such reade
texts. Like knowledge of the lexis and grammar of a language, this kind of knowledge from I
is part of what readers need to know to read successfully in a second language. give a
much
Reading Rate 1£a
the di
Fluent decoding depends on the reader having achieved what reading specialists many
call "automaticity," that is, the ability to convert most written language into mean- canne
ingful information so automatically that the reader does not have to think about the
language and can concentrate on combining the information obtained with back- ThE
ground knowledge to construct a meaning for the text. This requires the kinds of one
knowledge reviewed earlier-knowledge of the real world and knowledge of lexis, son
grammar, and text structure-but it also requires the skill of reading in meaningful we]
groups of words, sometimes called" chunking" and often described in terms of reading one
rate, the ability to decode so many words per minute. Fluent decoding is thus both can
see
rapid and accurate decoding, because the human brain cannot acquire information fan
from language that it does not understand or from language that is being processed can
too slowly. In practice, the two are interrelated, given the way in which the human aga
memory system works. The system has three parts: sensory store, short-term memory, one
and long-term memory (Stevick, 1976). Sensory store merely records the visual image
of the script or print the reader is reading. Short-term memory converts this image When
into meaningful information. Since it holds only 5 to 7 units at a time, for efficient know
reading these units must be reasonably large and meaningful. The units cannot be By
individual words, that, similarly, have little meaning by themselves. Words do not, (relati
as common sense suggests, give meaning to sentences so much as sentences give
meaning to words. Consider these sentences: Itv
ran
There is water in the well.
He was sick but he is well now. l.
He plays football wel/. 2.
He saw tears well up in her eyes. 3.
"Well, I don't know," he said. 4.
5.
By itself, the word well has many possible meanings-and therefore no clear 6.
meaning-but when combined with other words, it takes on clear meanings. To 7.
read for meaning, then, a reader must bring meaningful groups of words into short- 8.
term memory. Letter-by-letter or word-by-word reading fill short-term memory with 9.
meaningless units; no meaningful information gets through to combine with the 10.
reader's background knowledge for placement in long-term memory (where knowl-
edge is stored in the form of concepts or ideas, not words). Automaticity is thus a prod- The fi
uct of both knowledge of language and skill in processing language in written form. releva
 31. READING IN A SECOND LANGUAGE 569

:ling com- Reading and Background Knowledge (Content Discourses)

aboration,
Successful reading begins with fluent decoding, but this must be accompanied by
the reader's construction of a meaning for the text (commonly referred to as read-
yntactic ing comprehension), which goes well beyond decoding. Every written text provides
the pic- information for the reader, but the meaning of the text must be determined by a
.ctors as reader who can relate that information to some relevant body of knowledge. To make
sense of the new information provided by a text, or written discourse, a reader must
have some knowledge of what the discourse is about-its content-to which to relate
sests that that new information. The reader's brain is not an empty container to be filled with
r 1992; Ri-
meaning from the text. The brain is full of knowledge in the form of schemata, which
discourse collectively add up to "a picture of the world," according to Frank Smith (1975), that
lend such readers carry around in their heads. Therefore the brain relates new information taken
nowledgs from the text to the much larger body of knowledge it already has to make sense of or
age. give a meaning to the text as a whole. As Smith says "what the brain tells the eyes" is
much more important that "what the eyes tell the brain."
If a reader cannot determine what a text is about, that reader cannot comprehend
the discourse it contains even if he or she can decode the text perfectly. For example,
pecialists many people can decode each of the sentences in the following "opaque" text but
to mean- cannot even say what the text is about, let alone what it means:
about the
ith back- The procedure is quite simple. First, you arrange the items in separate piles. Of course,
kinds of one pile may be sufficient depending on how much there is to do. If you have to go
~of lexis, somewhere else due to lack of facilities, that is the next step; otherwise, you are pretty
!aningful well set. It is important not to overdo things. That is, it is better to do too few things at
)freading once than too many. In the short run, this may not seem important but complications
hus both can easily arise. A mistake can be expensive as well. At first, the whole procedure will
ormation seem complicated. Soon, however, it will become just another facet of life. It is difficult to
rocessed foresee any end to the necessity for this task in the immediate future, but then, one never
can tell. After the procedure is completed, one arranges the materials into different groups
2 human
again. Then they can be put into their appropriate places. Eventually they will be used
memory, once more and the whole cycle will then have to be repeated. However, that is part of life.
al image
is image When told that the subject is washing clothes, however, most readers can apply that
efficient knowledge to a second reading of the text and suddenly understand it perfectly.
mnot be By contrast, most American readers have no trouble comprehending the following
; do not, (relatively) normal text and answering the comprehension questions that follow:
ces give
It was the day of the big party. Mary wondered if Johnny would like a kite. She
ran to her bedroom, picked up her piggy bankand shook it. There was no sound!

1. Does this story take place in the past, present, or future?

2. What did Mary wonder?
3. What does the word would signal?
4. What is a kite?
5. What is a piggy bank?
lO clear 6. What kind of party is this text about?
ngs. To 7. Are Mary and Johnny adults or children?
o short- 8. How is the kite related to the party?
iry with 9. Why did Mary shake her piggy bank?
vith the 10. Mary has a big problem; what is it?
knowl-
a prod- The first five questions can be answered by decoding alone: The text provides the
nform. relevant information (as long as the reader knows the meaning of the words and
570 ESKEY

structures in English and recognizes the text as a story); but to answer the second five oppn
questions (simple enough for most native speakers but more problematical for non- fu1 tc
native speakers who may not have a Western kids' birthday party schema), the reader thou!
must bring cultural knowledge to the text, which provides no direct information on of tee
these topics. In other words, to comprehend even this simple story, a reader must, it grc
simultaneously, engage in bottom-up decoding and top-down interpretation of the unfoi
text to construct a plausible meaning for it, a process called parallel processing. of me
Thus, reading as a psycholinguistic process, when performed successfully, entails that 1
both rapid and accurate decoding and the construction of meaning based on prior Soka.
knowledge. Second language readers often have problems with both processes. cours
Quar
Reading as Sociocultural Practice as a E
discu
It is now generally understood that literacy varies from culture to culture. That a cru
is, the members of different cultures do different kinds of reading (and writing) for
different purposes. Most human beings learn to speak at least one language and
therefore use language to communicate with others, but people must be taught to kir
read and may never learn to do so. As human beings we have what could fairly int
be called a biological instinct to learn to speak, but we must be taught to read in ap
some particular culture that employs written language for some particular purposes. ma
Thus becoming literate means not only acquiring the kinds of skills and knowledge me
already discussed; it also means being enculturated (to the reader's own culture) or
acculturated (to another culture) in a kind of apprenticeship, which Smith (1988) has Such
compared to joining a club-the literacy club-composed of those who read and write plem
in some particular culture. prop
forei,
Ll/L2: Phonology and Orthography. Given the variety of writing systems, even mom
decoding may differ crosslinguistically. Among others, Koda (1989) and Haynes and impe
Carr (1990) have determined, for example, that students of English as a Second Lan- lingu
guage (ESL) who have learned to read in a language that does not employ an alphabetic man-
writing system often have problems in decoding English texts. more
that:
Reading and Background Knowledge (Cultural Discourses). At higher levels, little
most writers write for a culturally similar audience of readers and thus assume to co
that these readers share a common knowledge base and a common value system,
but common knowledge and values vary across groups (e.g., across cultures and
across classes and ethnic enclaves even within a single complex culture). Writers
also produce the kinds of discourse that have evolved naturally within their cul- OJ
tures (e.g., sonnets or haiku, personal essays or research papers), which may not be allth
familiar to the second language reader. Although certain kinds of disciplinary knowl- even
edge (like scientific knowledge) can reasonably be described as universal, second a liI(
language readers frequently encounter topics and attitudes in their reading (as in With
the birthday party story) that are new or strange to them and interfere with com- read,
prehension. Ever
Thus learning to read in a second language not only entails mastering a new lan- beha
guage in its written form, but also learning to engage in a new set of social practices becoi
that may conflict with those the reader is used to. lot tc
a pre
Critical Literacy. During the past 10 or 15 years, a body of work has emerged kind:
in the field dealing with what is sometimes called "the politics of literacy," or, more they
grandly, "critical literacy," which addresses language teaching in relation to various their
sociopolitical concerns. The patron saint of this work is Paolo Freire who, in his ear- readr
liest writings, made the valid point that in teaching so-called world languages to him!
 31. READING IN A SECOND LANGUAGE 571

second five oppressed peoples, second language teachers should teach these languages as use-
cal for non- ful tools to be used in overcoming their oppression. Although some of this work is
,the reader thought-provoking and a useful reminder that language teaching, like other kinds
rrnation on of teaching, does involve sociopolitical issues (Benesch, 1993; Gee, 1990), much of
.ader must, it grossly exaggerates the political dimension of language teaching. Much of it also,
ition of the unfortunately, reflects the kind of political correctness that has made a laughingstock
,mg. of many English departments at U.S. universities. Based on post-everything analyses
ully, entails that have largely been laughed out of the hard sciences as a result of physicist Alan
ed on prior Sakal's famous hoax (in which he submitted a transparent parody of postmodern dis-
:esses. course, "Transgressing the Boundaries-Toward a Transformative Hermeneutics of
Quantum Gravity," to a major postmodern journal that promptly published the piece
as a serious contribution to scholarship (Sokal, 1996; see Weinberg, 1996, for a lively
discussion of this event), this work attempts to place language teaching in the van of
11ture. Tha t a crusade for social justice. As Kaplan and Baldauf observe, however,
writing) for
19uage and ... one must be careful not to replace one kind of exploitation of minorities with another
e taught to kind, or to replace one existing minority with a new minority created by the process
.ould fairly intended to redress injustice ... This is in fact what Friere seems to recommend in his
: to read in approach to the empowerment of minorities; he suggests turning the minority into a
r purposes. majority and creating a new minority out of the present majority so that the new minority
knowledge may be exploited by the old minority. (Kaplan & Baldauf [r., 1997, p. 81)
culture) or
1 (1988) has Such recommendations are, in any case, rarely feasible or in any danger of being im-
d and write plemented by people with real political power, and this work is full of Canute-like
proposals for turning back the sea of history. The teaching of English as a second or
foreign language is a favorite target, English being guilty of having acquired "hege-
items, even monic" status, and ESL teachers are frequently accused of aiding and abetting an
-Iaynes and imperialist plot by the United States and United Kingdom to take over the world by
econd Lan- linguistic means, destroying other languages in the process (Phillipson, 1992). Since
l alphabetic many in the language teaching profession seem committed to the simplistic credo "the
more cultures and languages, the better" (as opposed to recognizing the obvious-
that multilingualism and multiculturalism can be either a blessing or a curse, or a
;her levels, little of both), this work will always find an avid readership, but it seems to have little
LUS assume to contribute to the improvement of second language teaching.
lue system,
.iltures and Reading as Individual Behavior
re). Writers
n their cul- Of course, no human reader is ever just a generic text processor or a simple clone of
may not be all the other members of some literacy club (or even a combination of the two), though
iary knowl- every kind of reading is strongly constrained by the nature of the reader's brain as
sal, second a linguistic processing device and by the reader's social and cultural experience.
ding (as in Within these constraints, however, readers differ in what they read, how much they
with com- read, how well they read, and how much they depend on or care about reading.
Every reader is, in short, an individual whose attitudes toward reading and reading
a new Ian- behavior are, to a considerable extent, idiosyncratic and unpredictable. Moreover, to
al practices become a skillful reader, a reader must read a lot (just as a swimmer must swim a
lot to become a skillful swimmer). Thus engaging in extensive reading behavior is
a prerequisite for developing reading skills, especially at the level required for most
as emerged kinds of formal education, and readers are most likely to engage in such behavior if
'I," or, more they have access to texts that are interesting to them as individuals and relevant to
l to various their particular needs. Reading from this point of view consists of every individual
,in his ear- reader developing a reading habit over time by reading texts of interest and value to
nguages to him or herself and reading those texts extensively.
572 ESKEY

Case Studies and Reading Protocols. Case studies are one means of investigat- det:
ing readers as individuals, and the arguments that Neuman and McCormick (1995) bas
advance for this kind of single subject research appear to apply with equal force to beli
first and second language reading. Cho and Krashens (1994) study of a Korean girl skil
who became enamored of the series of books about Sweet Valley High provides a well-
known example of such work. A second means of investigating readers as individuals 1
is protocol analyses, which requires direct interaction with readers. Bernhardt (1991) i
makes a strong case for the use of protocols in researching, teaching, and testing L2 a
reading on the grounds that there is no way of predicting what specific problems a s
given group of second language readers will have in comprehending particular texts. a

1
Extensive Reading. A major topic in the field of teaching L2 reading is extensive c
reading, many versions of which allow individual readers to select their own texts.
r
Day and Bamford (1998) is an excellent overview; Krashen (1993) makes the case for 1<
this approach to instruction; and Elley and Mangubhai (1983) and Robb and Susser t
(1989) provide reviews of successful extensive reading programs. Since this is an 1<

approach to the teaching of reading, it will be discussed in more detail next.

a
Evaluation of Reading r
c
Reading is the hardest language skill to assess because so much depends on what 1
is being read by whom. Beyond a certain minimal competence, there is no general c
proficiency in reading, every reader being more proficient at reading some texts than
others. Thus any passage selected for testing will favor some readers and disadvantage
j
others, since no two readers have exactly the same proficiency in language or exactly
the same funds of knowledge. In testing reading comprehension, it is also difficult adc
In 1
to determine what kinds of questions any reader who understands a text should be
by :
able to answer-that is, what constitutes reading comprehension for particular texts.
(NI
In practice, of course, educators do attempt to assess reading skills, and Alderson
of c
(2000) provides a recent and comprehensive review of the most compelling work
in this area. Significant variables identified include question types (e.g., multiple- real
choice versus open-ended); language effects (e.g., the language of assessment-L1 or
(2)
L2); and second language proficiency (e.g., more proficient, experienced versus less the
experienced, beginning) (Wolf, 1993). In summarizing this literature, Carrell (2001) apF
ma:
observes:
all (-
What can safely be concluded from the results of research on the assessment of second lan-
guage reading is caution in the interpretation of findings utilizing reading tests .... Perfor-
mance on reading assessment measures is dependent upon many factors, including the
assessment task type, the language of the assessment, the reading texts, and so on. There-
fore, researchers often recommend multiple measures for testing reading comprehension.
(Shohamy, 1984)

IMPLICATIONS FOR TEACHING

Thi
Reading Research and the Teaching of Reading dist
of n
A major issue in the field of L2 teaching is the relationship between such teaching oft
and research on language acquisition. Currently, many teachers, and many more re- 1
searchers, seem to take it for granted that good teaching practices can and should be trib
derived directly from research. The only problems with this notion that they recog- is n
nize are identifying which research is best and filling in whatever gaps may exist. also
The road from research to practice is assumed to be a one-way thoroughfare with no Alli
 3 I. READI G IN A SECONDLANGUAGE 573

nvestigat- detours, and teachers, who often take other routes, are frequently condemned for not
Lick(1995) basing their teaching on "scientific" research results. These are dangerously simplistic
31force to beliefs, because teaching and research call for very different kinds of knowledge and
mean girl skills. As Ellis (1998) observes:
Ies a well-
.dividuals
Teachers operate in classrooms where they need to make instantaneous decisions regard-
rdt (1991) ing what and how to teach. Researchers, more often than not, work in universities, where
testing L2 a system of rewards prizes rigorous contributions to a theoretical understanding of is-
roblems a sues. Teachers require and seek to develop practical knowledge; researchers endeavor to
ular texts. advance technical knowledge ....

extensive Technical knowledge is acquired deliberately either by reflecting deeply about the object
of inquiry or by investigating it empirically, involving the use of a well-defined set of
rwn texts.
procedures for ensuring the validity and reliability of the knowledge obtained. Technical
e case for knowledge is general in nature; that is, it takes the form of statements that can be applied
.id Susser to many particular cases. For this reason, it cannot easily be applied off-the-shelf in the
this is an kind of rapid decision making needed in day-to-day living ....

In contrast, practical knowledge is implicit and intuitive. Individuals are generally not
aware of what they practically know .... [It] is acquired through actual experience by
means of procedures that are only poorly understood. Similarly, it is fully expressible
; on what only in practice, although it may be possible, through reflection, to codify aspects of it.
The great advantage of practical knowledge is that it is proceduraJized and thus can be
) general
drawn on rapidly and efficiently to handle particular cases. (pp. 39-40).
exts than
:ivantage
rr exactly As a major case in point, in the United States reading policy makers have recently
, difficult adopted a very strong version of this teaching-derived-directly-from-research fallacy.
hould be In 1999, a committee of researchers dubbed "The National Reading Panel." convened
Jar texts. by the director of The National Institute of Child Health and Human Development
<\lderson (NICHD) at the behest of Congress, submitted a report intended to assess the validity
ng work of current reading research and the implications of that research for the teaching of
nultiple- reading. This report attempted to establish (1) what constitutes valid reading research,
It-11 or (2) what research is most relevant to teaching, and (3) what the implications are of
rsus less the research selected for "best practices" in the teaching of reading. As noted in the
.n (2001) appended "Minority View," the findings of the panel are (not surprisingly, given the
magnitude of the charge and the few months the panel had to complete it) suspect on
all counts but especially on the question of teaching:
id lan-
"erfor- As a body made up mostly of university professors ... its members were not qualified to be
ng the the sole judges of the "readiness for implementation in the classroom" of their findings
There- or whether the findings could be "used immediately by parents, teachers, and other
nsion. educational audiences." Their concern, as scientists, was whether or not a particular line
of instruction was clearly enough defined and whether the evidence of its experimental
success was strong. (Minority View, P: 2)

This has not, however, prevented the widespread adoption of the report by U.S. school
districts as the last word on reading and the teaching of reading, much to the detriment
of reading teachers whose experience has led them to different conclusions from those
eaching of the report.
nore re- The real issue here is not whether teachers or researchers know best, but what con-
.ould be tributes most to success in the teaching of reading-or any other subject. The answer
y recog- is neither research-based practices nor particular approaches, methods, and materi-
ry eccis]. als. The answer is good teaching. Speaking for himself and his associates, Richard
with no Allington (2000) writes:
574 ESKEY

A series of studies have confirmed what was probably obvious from the beginning. regu
Good teachers, effective teachers, matter much more than particular curriculum materials, wha
pedagogical approaches, or "proven programs." It has become clear that investing in
good teaching-whether through making sound hiring decisions or planning effective Mai
professional development-is the most "research-based" strategy available. If we truly
hope to attain the goal of "no child left behind," we must focus on creating a substantially
The
large number of effective, expert teachers .... Effective teachers manage to produce better
achievement regardless of which curriculum materials, pedagogical approach, or reading the I
program they use. (pp. 742-743) read
texts
and
Thus the suggestions for teaching that follow are not offered as magic formulas
qual
guaranteed to produce success, but as guidelines that might be helpful to good teach-
but
ers, who will know how to adapt them in practice for the particular students in their
particular classrooms. teacl
as IT
to re
Classroom Procedures

The problem for most teachers of reading in any language is that reading does not Exh
generate any product that a teacher can see or hear. Reading is an invisible process. (It is
Asr
therefore much like listening, but a teacher can ask students to perform tasks while
agot
listening-for example, giving students a dictation). Most teachers take the process of IE
for granted and go directly to the creation of a related product-for example, asking
Unli
students to answer comprehension questions orally or in writing. These activities test
are I
reading but do not teach it, and this contributes little to improving any student's The
reading performance.
C
forrr
Intensive/Extensive Reading (Dre
subs
Historically, procedures for teaching reading have often been divided into procedures
Sam
for teaching intensive reading (working with small amounts of text in class to make
choi
various points about the nature of texts and the reading process) and procedures for
are}
teaching extensive reading (assigning whole texts to be read outside of class or in a
Sam
reading lab setting). These are useful categories for structuring programs, but they
and
do not shed much light on the purpose of asking students to engage in either kind of
read
activity-that is, on how engaging in such activities can help them to become better
readers.
Fad
One good way of addressing this question is to turn the question upside down:
How do people learn to read a language? And, once they have learned to do it, how do Proe
they learn to read better? The answer to both questions is surprisingly Simple. People how
learn to read, and to read better, by reading. No one can teach someone else to read: is, OJ
The process is largely invisible and thus cannot be demonstrated, and it mainly occurs teac.
at the subconscious level and thus cannot be explained in any way that a reader could area
make conscious use a£.2 However, anyone can learn to do it, just as anyone can learn pub:
to draw or to sing at some minimal level of competence. Every normal human being
is capable of learning to read, given the right opportunity and guidance.
The reading teacher's job is thus not so much to teach a specific skill or content as
to get students reading and to keep them reading-that is, to find a way to motivate Thu:
them to read, and to facilitate their reading of whatever texts they have chosen to read ing
or been asked to read. cess:
mea
Motivation dicti
forn
Procedures for motivating students to read tend to focus on students as individuals or ing I
as members of particular groups. To engage in something as challenging as reading teac.
 31. READINGIN A SECONDLANGUAGE 575

>ginning. regularly in a second or foreign language, learners must be highly motivated, and
.iaterials, what motivates one reader or group of readers might not motivate another.
esting in
effective
we truly Matching Readers and Texts
.tantially
ce better
'reading
t The solution to this problem begins with locating appropriate texts, that is, texts that
the reader wants or needs to read. Since people learn to read, and to read better, by
reading, a major part ofthe reading teacher's job is to introduce students to appropriate
texts-texts at the right level linguistically and texts that are both interesting to them
. formulas and relevant to their particular needs-and to induce them to read such texts in
ood teach- quantity. For some students, it may be enough to make appropriate texts available,
its in their but for others, more guidance may be required. For the full range of students, the
teacher must create his or her version of the literacy club and find ways to persuade
as many students as possible to join and to become literate-that is, to read texts and
to respond to those texts in the ways that typical club members do.

does not Extensive Reading Programs

·cess. (It is
As noted earlier, extensive reading programs have recently emerged as a major ped-
sks while
agogical response to the problem of finding appropriate texts for particular groups
e process
of readers or for individuals, and for inducing them to read such texts in quantity.
le, asking
Unlike earlier attempts to incorporate extensive reading in curricula, these programs
vities test
are not mere components in a larger program that also involves intensive reading:
student's
The extensive reading is the program.
Often called Sustained Silent Reading (SSR) today, these programs come in various
forms. Under such additional acronyms as FVR (Free Voluntary Reading), DEAR
(Drop Everything and Read), and many others, all of them require L2 students to read
ocedures substantial amounts of L2 text. They differ with respect to degree of student choice:
to make Some allow students to choose their own materials; some allow students to make
lures for choices from prescribed reading lists; some assign readings; and various combinations
s or in a are possible. They also differ with respect to what is expected of participating students:
butthey Some require nothing more than the reading; some require summaries or book reports;
r kind of
and some require, more traditionally, exams or writing assignments based on the
te better
readings.

e down: Facilitation
how do Procedures for facilitating L2 reading tend to focus on teaching the reader (any reader)
. People
how to manage reading as a cognitive process more painlessly and efficiently-that
to read:
is, on making L2 reading as easy as possible for the learner. Most current work on the
y occurs
teaching of such reading takes this general approach, possibly because this is the one
er could area in which reading teachers can justify doing some direct teaching. It also gives
an learn
publishers something to publish, that is, materials for teaching strategies for reading.
n being
Teaching Cognitive Strategies
ntent as
iotivate Thus in addition to providing-or providing access to-appropriate materials, read-
to read ing teachers often teach cognitive strategies for reading, both for bottom-up pro-
cessing (e.g., reading at a reasonable rate-which, as noted, really means reading in
meaningful groups of words-and reading without stopping to look up words in the
dictionary) and for top-down processing (e.g., skimming a text before reading and
formulating specific questions that the text might be expected to answer). The follow-
'uals or ing checklist provides an overview of the kind of strategies reading teachers often
eading teach:
576 ESKEY

A CHECKLISTOF READING STRATEGIES: Cl

1. Prereading.
A. Bottom up, e.g., vocabulary building In
B. Top down, e.g., schema bulding to fac
C. Text attack, e.g., skimming comr:
2. While reading. and s
A. Bottom up, e.g., employing knowledge of morphology (any word with tempor Be
in it has something to do with time) teach
B. Top down, e.g., reading to find the answers to questions (SQ3R) must
3. Post reading Good
For example, asking questions based on purpose of text or asking questions that for th
call for critical reading norm
4. Follow-up as asl
For example, moving on to other texts on the same topic or to other language text r
modes (listening, speaking, writing) (1998
kind (
asmc
There is a substantial literature on the teaching of such strategies, both in the form the te
of resources for teachers and text materials for students. In one of the better texts for work
teachers, Anderson (1999),for example, identifies, discusses, and promotes eight use-
ful teaching strategies-activating background knowledge, cultivating vocabulary,
teaching for comprehension, increasing reading rate, verifying reading strategies,
evaluating progress, building motivation, and selecting appropriate materials-many 1. To tI
of which entail teaching students to employ specific cognitive strategies for success- com
fully decoding and interpreting texts. In fact, most of the current literature on the revis
teaching of second language reading is largely devoted to the teaching of such strate- 2. It hi
Rorr
gies (e.g., Aebersold & Field, 1997;Nuttall, 1996;Urquhart & Weir, 1998;Day, 1993). can 1
So, for that matter, is much of the literature on the teaching of first language reading
(e.g., Mikulecky, 1990).The first L2 textbook based on a psycholinguistic model and
the first to incorporate this approach was Reader's Choice by Silberstein, Dobson, and
Clarke (2002), which is now in its fourth edition, but many others have appeared
during the past 20 years (e.g., Zukowski-Faust, Johnson, & Templin, 2002; Rosen & Aebers.
Stoller, 1994;and Mikulecky, 1990). Alders:
Allingt
Anders
Benescl
Teaching Metacognitive Strategies Bensou
ResE
As interest in the teaching of strategies developed, many reading specialists and teach- Bernha
Nor
ers discovered that simply teaching students a list of cognitive strategies for read- Bernha:
ing did not help every student to become a better reader. As it happens, successful (Ed.
cognition, or thinking, must be directed and monitored by higher levels of cognition Lan,
called metacognition, or thinking about thinking. The issue is not just what strategies Bernha:
the]
can be used and how to use them, but when to use them and for what purpose. In Birch, B
dealing with real texts, the issue is always what problems the text is creating for the Carrell,
reader and what strategies he or she might employ to address and hopefully solve Carrell,
these problems. More recent work on reading incorporates these insights. Carrell,
texts
In support of this approach, a number of research studies suggest that teaching Carrell,
reading strategies can have positive effects on the reading performance of second 97-1
language learners. Carrell, Gajdusek, and Wise (1998)attempted to determine which Carrell,
of these studies met the criterion of teaching both cognitive and metacognitive strate- brid
Cho, K.
gies; they cite Carrell (1985) and Raymond (1993) as having fully met this criterion. acqL
Other studies of note include Carrell, Devine, and Eskey (1988), Kern (1989), and Clarke,
Zimmerman (1997)for the teaching of vocabulary. read
 31. READING IN A SECOND LANGUAGE 577

Classroom Procedures: A Final Note

In addition to motivating students, successful L2 reading teachers do their best
to facilitate reading for them. These two kinds of procedures are, it should be noted,
complementary, since students who enjoy reading are more likely to read successfully,
and students who read successfully are more likely to enjoy it.
withtempor Because no two classes are alike, however, it cannot be assumed that any reading
teacher can know in advance what his or her students' major problems will be. That
must be determined by interacting directly with a given group of readers as they read.
Good second language reading teachers create, as noted, a new kind of literacy club
-stions that for their students, sharing their own reading and responding to it as native speakers
normally do. They also read with their students, making use of such simple protocols
as asking students to paraphrase what they are reading or to speculate on where the
rlanguage text might be going in order to determine what their real problems are. Bernhardt
(1998) makes a compelling case for this, and reminds the profession that teaching any
kind of reading successfully most often requires a skillful and dedicated teacher as well
as motivated and competent students. There are no magic approaches or methods for
1 the form the teaching or learning of second language reading, but good teachers and students,
er texts for working together, sometimes get the job done successfully.
eight use-
ocabularv, NOTES
strategies,
LIs-many
l. To the great loss of the applied linguistics community, David Eskey passed away in October 2002 after
ir success- completing the first draft of this chapter. Thanks to Eleanor Black Eskey for materials used in the final
tre on the revisions of the manuscript.
ich strate- 2. It has been noted, however, by Birch (2002) and others that L2 readers whose L1 does not have a
)ay, 1993). Romanized alphabet can experience difficulties with orthographic and phonological correlations that
can be taught.
;e reading
rodel and
bson, and REFERENCES
appeared
Rosen & Aebersold, J., & Field, M. (1997). From reader to reading teacher. New York: Cambridge University Press.
Alderson, J. C. (2000). Assessing reading. Cambridge, UK: Cambridge University Press.
Allington, R. (2000). What really matters for struggling reader. London: Longman.
Anderson, N. J. (1999). Exploring second language reading: Issues and strategies. Boston, MA: Heinle & Heinle.
Benesch, S. (1993). ESL, ideology, and the politics of pragmatism. TESOL Quarterly, 27(4), 705-717.
Bensoussan, M., & Laufer, B. (1984). Lexical guessing in context in EFL reading comprehension. Journal of
Research in Reading, 7, 15-32.
nd teach- Bernhardt, E. (1991). Reading development in a second language: Theoretical, empirical, and classroom perspectives.
for read- Norwood, NJ: Ablex.
Bernhardt, E. B. (1998). Sociohistorical perspectives on language teaching in modern America. In H. Byrnes
uccessful (Ed.), Perspectives in research and scholarship in second language learning (pp, 39-57). New York: Modern
.ognition Language Association.
;trategies Bernhardt, E. B., & Kamil, M. L. (1995). Interpreting relationships between L1 and L2 reading: Consolidating
rpose. In the linguistic threshold and the linguistic interdependence hypotheses. Applied Linguistics, 16 (1), 15-34.
Birch, B. (2002). English L2 reading: Getting to the bottom. Mahwah, NJ: Lawrence Erlbaum Associates.
g for the Carrell, P. (1992). Awareness of text structure: Effects on recall. Language Learning, 42,1-20.
Ily solve Carrell, P. L. (1985). Facilitating ESL reading by teaching text structure. TESOL Quarterly, 19(4), 727-752.
Carrell, P. L. (2001). The influence of purpose for reading on second language reading: Reading procedural
teaching texts in ESL. Reading in a Foreign Language, 13(2). 220-251.
Carrell, P. L., Gajdusek, L., & Wise, T. (1998). Metacognition and EFL/ESL reading. Instructional Science, 26,
f second 97-112.
le which Carrell, P., Devine, J., & Eskey, D. (1988). lntemctioe approaches to second language reading. New York: Cam-
'e strate- bridge University Press.
'riterion, Cho, K.-S., & Krashen, S. (1994). Acquisition of vocabulary from the Sweet Valley Kids series: Adult ESL
acquisition. Journal of Reading, 37,662-667.
89), and
Clarke, M. (1980). The short circuit hypothesis of ESL reading: When language competence interferes with
reading performance. The Modem Language [ournal, 64, 114-124.
 578 ESKEY

Cummins, J. (1984). Bilingualism and special education: Issues in assessment and pedagogy. Clevedon, UK: Robb,
Multilingual Matters. Fo
Day, R, & Bamford, J. (1998). Extensive reading in the second language classroom. New York: Cambridge Rosen
University Press. Rume
Day, R (Ed.), (1993). New ways in teaching reading. Alexandria, VA: TES01. Bn
Elley, W., & Mangubhai, F. (1983). The impact of reading on second language learning. Reading Research AE
Quarterly 19, 53-67. Segal<
Ellis, N. (1998). Emergentism: Connectionism and language learning. Language Learning 48, 31-64. lar
Favreau, M., & Segalowitz, N. S. (1983). Automatic and controlled processes in the first and second language Shanr
reading of fluent bilinguals. Memory and Cognition, 11(6),565-574. 43'
Friere, P., & Macedo, D. (1987). Literacy: Reading the word and the world. South Hadley, MA: Bergin & Garvey. Shayv
Fries, C. (1945). Teaching and learning English as a foreign language. Ann Arbor: University of Michigan Press. (2C
Fry, E. (1981). Graphical literacy. Journal of Reading, 24, 383-389. P.l
Gee, J. (1990). Social linguistics and literacies. Bristol, PA: Falmer Press. Er:
Gee, J. (2000). Discourse and sociocultural studies in reading. In M. 1. Kamil, P. B. Mosenthal, P. D. Pearson, Shoha
& R. Barr (Eds.), Handbook of reading research, Vol III (pp. 195-208). Mahwah, NJ: Lawrence Erlbaum Lai
Associates. Silber:
Geva, E., Wade-Woolley, 1., & Shaney, M. (1993). The concurrent development of spelling and decoding in Pn
two different orthographies. Journal of Reading Behavior, 25, 383-406. Smith
Gilbert, P. (1993). Narrative as gendered social practice: In search of different story lines for language Smith
research. Linguistics and Education, 5, 211-218. Smith
Goodman, K. S. (1967). Reading as a psycholinguistic guessing game. Journal of the Reading Specialist, 4, Sokal,
126-135. Sac
Goodman, K. S. (1975). The reading process. In F. W. Gollash (Ed.), Language and literacy: The selected writings Stano'
of Kenneth Goodman (pp. 5-16). London: Routledge & Kegan Paul. op
Haynes, M., & Carr, T. (1990). Writing system background and second language reading: A component skills Stano'
analysis of English reading by native speaker-readers of Chinese. In J. C. Alderson & A. H. Urquhart P'l
(Eds.), Reading and its development: Component skills approaches (pp. 375-421). New York: Academic Press. Stano'
Heath, S. B. (1983). Ways with words: Language, life, and work in communities and classrooms. New York: rea
Cambridge University Press. NJ
Huckin, T., Haynes, M., & Coady, L (Eds.). (1993). Second language reading and vocabulary learning. Norwood, Stevie
NJ:Ablex. Street,
Hudson, T. (1982). The effects of induced schemata on the short circuit in L2 reading: Non-decoding factors Urqul
in L2 reading performance. Language Learning, 32,1-31. Lo
Huey, E. B. (1908). The psychology and pedagogtj of reading. With a review of the history of reading and writing and Wardl
of methods, texts, and hygiene in reading. New York: Macmillan. Weinb
Kamil, M. 1., & Lane, D. M. (1998). Researching the relationship between technology and literacy: An Wolf,
agenda for the 21st century. In D. Reinking, M. McKenna, 1. D. Labbo, & R Kieffer (Eds.), Handbook of Lar
literacy and technology: Transformations in a post-typographic world (pp. 323-342). Mahwah, NJ: Lawrence Zimm
Erlbaum Associates. pir
Karnil, M. 1., Intrator, S. M., & Kim, H. S. (2000). The effects of other technologies on literacy and literacy Zukov
learning. In M. 1. Karnil, P. B. Mosenthal, P. D. Pearson, & R. Barr (Eds.), Handbook of reading research, &J
Vol III (pp. 771-788). Mahwah, NJ: Lawrence Erlbaum Associates.
Kamil, M., Mosenthal, P., Pearson, P. & Barr, R. (2000). Handbook of reading research: Volume III. Mahwah, NJ:
Lawrence Erlbaum Associates.
Kaplan, R. B., & Baldauf [r., R. B. (1997). Language planning: From practice to theory. Clevedon, UK: Multilingual
Matters.
jil Kern, R. G. (1989). Second language reading strategy instruction: Its effects on comprehension and word
t )'1
inference ability. The Modern Language Journal, 73, 135-149.
Koda, K. (1989). The effects of transferred vocabulary knowledge on the development of L2 reading profi-
ciency. Foreign Language Annals, 22, 529-540.
Krashen, S. (1993). The power of reading. Englewood, CO: Libraries Unlimited,
McFarland, K. P., & Stansell.]. C. (1993). Historical perspectives. In 1. Patterson, C. M. Santa, K. G. Short, &
K. Smith (Eds.). Teachers are researchers: Reflection and action (pp. 12-18). Portsmouth, NH: Heinemann.
Mikulecky, B. (1990). A short course in teaching reading skills. Reading, MA: Addison-Wesley.
Neuman, S. B., & McCormick, S. (Eds.). (1995). Single subject experimental research: Applications for literacy.
Newark, DE: International Reading Association.
Nuttall, C. (1996). Teaching reading skills in a foreign language. Oxford: Heinemann.
Phillipson, R. (1992). Linguistic imperialism. Oxford, UK: Oxford University Press.
Pinker, S. (1994). The language instinct. New York: HarperCollins.
Pressley, M., & Afflerbach, P. (1995). Verbal protocols of reading: The nature of constructively responsive reading.
Hillsdale, NJ: Lawrence Erlbaum Associates.
Raymond, P. M. (1993). The effects of structure strategy training on the recall of expository prose for
university students reading French as a second language. The Modem Language Journal, 77(4), 445-458.
Riley, G. (993). A story structure approach to narrative text comprehension. Modern Language Journal, 77,
417-430.
 31. READINGIN A SECONDLANGUAGE 579

:::Ievedon, UK: Robb, T. N., & Susser, B. (1989). Extensive reading versus skills building in an EFL context. Reading in a
Foreign Language, 5, 239-25l.
rk: Cambridge Rosen, N., & Stoller, F. (1994). Javier arrives in the U.S. Englewood Cliffs, NJ: Prentice-Hall.
Rumelhart. D. E. (1980). Schemata: The building blocks of cognition. In R J. Spiro, B. C. Bruce, & W. F.
Brewer (Eds.), Theoretical issues in reading comprehension (pp, 33-35). Hillsdale, NJ: Lawrence Erlbaum
,ading Research Associates.
Segalowitz, S., Segalowitz, N., & Wood, A (1998). Assessing the development of automaticity in second
31-64. language word recognition. Applied Psycholinguistics, 19, 53-67.
:ond language Shannon, P. (1996). Poverty, literacy, and politics: Living in the USA Journal of Literacy Research, 28, 430-
439.
rgin & Garvey. Shaywitz, B. A, Pugh, K. R., Jenner, A R, Fulbright, R K., Fletcher, J. M., Gore, J. c, & Shaywitz, S. E.
lichigan Press. (2000). The neurobiology of reading and reading disability (dyslexia). In M. L. Kamil. P. B. Mosenthal.
P. D. Pearson, & R Barr (Eds.), Handbook of reading research (Vol. 3, pp. 229-249). Mahwah, NJ: Lawrence
Erlbaum Associates.
P. D. Pearson, Shohamy, E. (1984). Does the testing method make a difference? The case of reading comprehension.
ence Erlbaum Language Testing, 1, 147-170.
Silberstein, S., Dobson, B., & Clarke, M. (2002). Reader's choice (4th ed.). AIm Arbor: University of Michigan
d decoding in Press.
Smith, F. (1973). Psycholinguistics and reading. New York: Holt, Rinehart & Winston.
for language Smith, F. (1975). Comprehension and learning. New York: Holt, Rinehart & Winston.
Smith, F. (1988). Joining the literacy club. Portsmouth, NH: Heinneman.
5 Specialist, 4, Sokal. A (1996). Transgressing the boundaries: Toward a trans formative hermeneutics of quantum gravity.
Social Text, 14, 217-252.
lected writings Stanovich, K. E. (1980). Toward an interactive-compensatory model of individual differences in the devel-
opment of reading fluency. Reading Research Quarterly, 16, 32-7l.
tponent skills Stanovich, K. E. (1991). Word recognition: Changing perspectives. In R Barr, M. L. Kamil. P. Mosenthal, &
H. Urquhart P. D. Pears en (Eds.), Handbook of reading research, (Vo!. 2, pp. 418-452). ew York: Longman.
sdemic Press. Stanovich, K. E. (1992). Speculations on the causes and consequences of individual differences in early
>. New York: reading acquisition. In P. Gough, L. Ehri, & R Treiman (Eds.). Reading acquisition (pp, 307-342). Hillsdale,
NJ: Lawrence Erlbaum Associates.
tg. Norwood, Stevick, E. (1976). Memory, meaning, and method. Rowley, MA: Newbury House.
Street, B. (1984). Literacy in theory and practice. Cambridge, UK: Cambridge University Press.
oding factors Urquhart, A. H., & Weir, C. J. (1998). Reading in a second language: Process, product, and practice. New York:
Longman.
d writing and Ward haugh, R (1977). The contexts of language. Cambridge, MA: Newbury House.
Weinberg, S. (1996). Sokal's hoax. The New York Review of Books, Vol. XLIII, ml111ber13 (August 8, 1996), 11-15.
literacy: An Wolf, D. (1993). A comparison of assessment tasks used to measure FL reading comprehension. Modern
, Handbook of Language Journal, 77, 473-489.
JJ: Lawrence Zimmerman, C. B. (1997). Do reading and interactive vocabulary instruction make a difference?: An em-
pirical study. TESOL Quarterly, 31,121-140.
and literacy Zukowski-Faust, J., Johnston, S., & Templin, E. (2002). Steps to academic reading (3rd ed.). Boston, MA: Heinle
ling research, & Heinle.

ifahwah,NJ:

vIultilingual

n and word

ading profi-

G.Short,&
einemann.

iiue reading.

1 prose for
),445-458.
[ournal, 77,