ArticlePDF Available
How Do Languages Change?
Höskuldur Thráinsson
University of Iceland
hoski@hi.is
https://uni.hi.is/hoski/
Outline of the talk
Introduction: The basic issues
Some distinctions and a brief overview:
1. change vs. diffusion
2. I-language and E-language
More distinctions and a continuing overview:
1. apparent-time vs. real-time
2. age grading vs. lifetime change
Ongoing research projects to study these issues:
1. Evidence for “linguistic change in apparent-time”
2. Evidence for “linguistic change in real-time”
Concluding remarks
Lund University,
September 6, 2012 2
Höskuldur Thráinsson
University of Iceland
Introduction
The main question: “How Do Languages Change?”
An illustrative example: Changes in the short vowel system of
Old Icelandic (OI, see e.g. HreinnBenediktsson (ed.) 1972;
2002:56, 60):
12th century: 13th century:
front back front back
unrd rd unrd rd unrd rd unrd rd
high i y u high i y u
mid e ¿ o mid e ¿ o
low () a Ä low a
Common belief: “This was a simplification of the system.”
Questions: How did this happen? Who “simplified the system”?
Lund University,
September 6, 2012 3
Höskuldur Thráinsson
University of Iceland
Introduction, 2
Hreinn’s comments on the merger (2002:60):
“This change took place at the end of the twelfth century or in
the beginning of the thirteenth, probably at somewhat
different times in the different varieties of Icelandic.”
Unanswered questions (and not commonly asked in
traditional historical linguistics: How did this happen?:
Three (?) possibilities:
1. OI speakers had the /¿ : Ä/ distinction in their language in the
(late) 12th century but after 1200 they (gradually?) lost it.
2. Shortly before 1200 or later new generations emerged who
did not have the /¿ : Ä/ distinction although their parents did.
3. The sound system just changed, “simplified itself”.
Lund University,
September 6, 2012 4
Höskuldur Thráinsson
University of Iceland
Introduction, 3
Assuming that languages do not change unless they are
being used, there are actually only two possibilities:
A Individual speakers “change their language” (always/
typically/most of the time ... unconsciously).
B Languages change when new generations acquire “a
different language/grammar” (“imperfect transmission”)
and hence speak differently.
Not necessarily mutually exclusive possibilities:
It could be that certain types of change occur as a
combination of A and B.
Maybe A holds for certain types of change (e.g. changes in
the lexicon/vocabulary) but B for other types of change
(phonological?, syntactic?)
Lund University,
September 6, 2012 5
Höskuldur Thráinsson
University of Iceland
Introduction, 4
More questions:
Why does traditional historical linguistics rarely consider
possibilities A and B, i.e. when investigating “old” linguistic
changes (like the “simplification” of the OI vowel system)?
Answer:
We typically do not have the appropriate data to decide
between the two possibilities when investigating changes in
older stages of a given language.
How can we the investigate the role/involvement/nature of
the two possibilities, A and B?
Answer:
By investigating current linguistic changes and try to
determine how they happen.
Lund University,
September 6, 2012 6
Höskuldur Thráinsson
University of Iceland
Introduction, 5
One more question and the most important one:
Why is it important
to learn more about possibilities A and B?
Answer:
It may increase our understanding of the nature of
linguistic change and hence the nature of human
language.
Lund University,
September 6, 2012 7
Höskuldur Thráinsson
University of Iceland
Some Distinctions and a Brief Overview
1. Change vs. Diffusion:
Some linguists (e.g. Hale 2007) make a sharp distinction
between (actual) change (i.e., the original “innovation”, which
may typically (always?) be the result of “imperfect
transmission”) and its diffusion (i.e., how/why the innovation
spreads to new generations/other speakers ...).
Sociolinguists are typically interested in diffusion (in Hale’s
sense), i.e. how and why certain variants spread throughout
the linguistic community, whereas generative linguists tend to
be more interested in the innovation (Hale’s change) and
how/why it has come about.
This is related to the distinction between I-language and E-
language.
Lund University,
September 6, 2012 8
Höskuldur Thráinsson
University of Iceland
Some Distinctions ..., 2
2. I-language vs. E-language:
I-language is the internal(ized)/individual language that
the speaker acquires, his/her knowledge of the
language.
By contrast, E-language (E for external(ized)) is pretty
much everything else we mean by language (e.g., when
we say: “vatn is not a word in the English language.”
Chomsky (1986 and later, see also Isac and Reiss 2008
and many others) has argued that I-language is/should
be main object of study of linguistic theory.
Lund University,
September 6, 2012 9
Höskuldur Thráinsson
University of Iceland
Some Distinctions ..., 3
A (slight) reformulation of a previous question:
Does or can linguistic change involve the change of the
language (“linguistic knowledge”) speakers have acquired, i.e.
their I-language (cf. possibility A on slide 5 above), or is
linguistic change proper only involved when a speaker of a
new generation has a different I-language than members of
previous generations because of some “innovation” or
“imperfect transmission” (cf. possibility B on slide 5 above).
Different opinions on this, cf. the overview below.
Lund University,
September 6, 2012 10
Höskuldur Thráinsson
University of Iceland
Some Distinctions ..., 4
A bit of history (my summary and interpretation):
Halle 1962 (long before Chomsky’s introduction of the distinction
between I-language and E-language even before Chomskys
distinction between competence and performance, e.g. 1965):
Adults change their language to some extent but not the actual
linguistic system (i.e., they may change their use/usage in various
ways). Children then interpret these changes as evidence for a
different underlying system. Then we have a change in the linguistic
system with a new generation.
Labov 1963, 1966 and later: Assumed that the linguistic knowledge
was relatively stable after the acquisition period and introduced
the concept of apparent-time to describe possible differences
between groups of speakers (cf. below).
Weinreich, Labov and Herzog (WLH) 1968: Emphasized the
relationship between linguistic change and variation. The believed
that diffusion of variants depended on sociological factors.
Lund University,
September 6, 2012 11
Höskuldur Thráinsson
University of Iceland
Some Distinctions ..., 5
A bit of history, contd.:
Andersen 1973 believed that WLH had not paid enough attention to the
origin of linguistic change and wanted to distinguish between rule
acquisition based on (incomplete) data (which may involve
reinterpretation or misinterpretation) and rule generalization which
extends the rule to new cases. In both instances we have a change but of
different nature (Anderson used the terms abductive and deductive
change ).
Lightfoot (1979 and later) emphasizes (like Halle 1962 and Andersen
1973) the role of children acquiring the language as instigators of change.
Hale (2007), too, sees language acquisition as the locus of linguistic
change and claims that change “does not take any time” and thus there is
no such thing as a “change in progress”. Change occurs when the child
misinterprets” the data, but the diffusion of change to the linguistic
community may take time. New generations adopt the innovation (it is in
their input). Adults can also “learn a new dialect”, e.g. the standard
dialect, but that is not really “linguistic change”.
Lund University,
September 6, 2012 12
Höskuldur Thráinsson
University of Iceland
Some Distinctions ..., 6
An interim conclusion:
Many emphasize the role of children acquiring language as
“instigators” of change, maintaining that this is where chance
occurs (innovations). A diagram illustrating this (in the spirit of
Hale 2007 and others, e.g. Margrét Guðmundsdóttir 2008):
Linguistic change
Mother’s I-lang., IL-1
Mother’s output (performance), OP-1 Child’s I-lang., IL-2
Child’s
(mis-)interpretation
Child’s output, OP-2
Lund University,
September 6, 2012 13
Höskuldur Thráinsson
University of Iceland
Some Distinctions ..., 7
More questions:
If we assume that the locus of linguistic change is typically in
language acquisition, does it necessarily follow that there can
be no changes in the linguistic system/competence/
knowledge/I-language ... of adults?
Is linguistic change in adults restricted by the nature of our
brain and how it matures? Does this have something to do
with the (alleged) lack of “plasticity” of the adult brain, cf.
ideas about the critical period in language acquisition (usually
traced back to Lenneberg 1967)?
Lund University,
September 6, 2012 14
Höskuldur Thráinsson
University of Iceland
Some Distinctions ..., 8
Different opinions:
Anderson abd Lightfoot (2002:209): “Whatever is in place by
puberty is what we are “stuck with”.”
Hale (2007, especially pp. 4445)): Changes in an adult’s
output may look like they derive from a change in the adult’s
grammar when what has in fact happened is that the adult
has acquired an additional grammar (e.g. a standard dialect
grammar) while keeping the original grammar “intact”.
Increased “use” of the new grammar, typically under certain
sociolinguistic conditions.
Grace (1969:105, acc. to Hale): Adults are capable of making
only certain kinds of changes in their grammars, but some
kinds of change are “most profitably conceptualized” as
changes in the grammars of adults rather than as
“restructuring” by children.
Lund University,
September 6, 2012 15
Höskuldur Thráinsson
University of Iceland
Some Distinctions ..., 9
Summary of some claims:
Many generative linguists (e.g. Hale 2007) want to distinguish
between change (the actual innovation) and diffusion.
Sociolinguistic factors undoubtedly affect diffusion but the
role of linguistic factors is debated: “I believe that diffusion is
a highly unconstrained processi.e., that any possible
‘change’ could just as easily diffuse under the proper
sociolinguistic conditions for diffusion” (Hale 2007:39).
Changes in linguistic behavior caused by diffusion are
reversible”, cf. that speakers can often revert to their native
phonological dialect even if they have mastered a standard
one (at least to some extent; see also Hale 2007:4445).
Some sociolinguists have, however, emphasized linguistic
factors in diffusion, e.g. in contact situations between
dialects (cf. Trudgill 1986, Kristján and Höskuldur 2003).
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 16
Some Distinctions ..., 10
One more question:
What kind of evidence could help us decide whether
adults (can) change their grammar (their I-language/
linguistic knowledge/linguistic system ...) as opposed
to, say, just learning a new grammar (to some
extent, e.g. a standard dialect or some other dialect)
that they may “use” under certain circumstances?
An answer:
Studying the “linguistic behavior” of the same speaker(s)
over a long period of time.
A problem:
The distinctions we have been considering may be too
“crude”. So here are some additional ones.
Lund University,
September 6, 2012 17
Höskuldur Thráinsson
University of Iceland
More Distinctions and a Continuing Overview
“Apparent-time” and “real-time” (originally from Labov):
Figure 1: Mean occurrence/acceptance of some linguistic phenomenon,
classified by age groups
Interpretation: No difference between age groups = Stability?
Lund University,
September 6, 2012 18
Höskuldur Thráinsson
University of Iceland
More Distinctions ..., 2
Labov’s ideas, contd.:
Figure 2: Mean occurrence/acceptance of some linguistic phenomenon,
classified by age groups
Three possible interpretations: age grading, lifetime
change, apparent-time (meaning what?)
Lund University,
September 6, 2012 19
Höskuldur Thráinsson
University of Iceland
More Distinctions ..., 3
Possible interpretations of figures like 1 and 2 (cf. e.g.
Labov 2001:83; Sankoff and Blondeau 2007:562563):
bars inter- individual linguistic
(lines): pretation: speakers: community:
1. even (level, cf. fig. 1) stability stable stable
2. even (cf. fig. 1) cont. change change changes
3a. uneven (sloping, cf. fig. 2) age grading change stable
3b. uneven (cf. fig. 2) lifetime change change changes
4. uneven (cf. fig. 2) apparent-time stable changes
Age grading: When a particular linguistic trait decreases or
incraases with age and this development repeats itself.
Lifetime change: When individuals change their language as they
grow older and this continues in the same direction with a new
generation.
Lund University,
September 6, 2012 20
Höskuldur Thráinsson
University of Iceland
More Distinctions ..., 4
(Another) Interim conclusion:
Bar charts like figs. 12 do not tell the whole story. They can
be interpreted in different ways (cf. slide 19). We cannot, for
instance, determine what is going on in the “linguistic
community” (or: whether the E-language is changing) by
doing a synchronic study comparing age groups and studying
bar charts of this kind.
So what can we do?
Do a comparable study of a given linguistic community,
preferably the same speakers, over a longer period of time, a
real-time study (longitudinal study).
Lund University,
September 6, 2012 21
Höskuldur Thráinsson
University of Iceland
More Distinctions ..., 5
Two kinds of real-time studies (cf. e.g. Sankoff and
Blondeau 2007:561):
In a trend study the language of comparable groups of
speakers is studied in a similar fashion two or more times
with several years in between the studies. That way the
direction of an possible change (or diffusion?) can be
determined.
In a panel study the language of the same group of speakers
(same individuals) is studied in a similar fashion two or more
times with several years in between the studies. That way
possible changes in the grammars of individual speakers can
be studied (or their acquisition of or changes in their use of
“different grammars”, cf. Hale 2007).
Lund University,
September 6, 2012 22
Höskuldur Thráinsson
University of Iceland
More Distinctions ..., 6
Once we have done a real-time (longitudinal) study, which
quetions could we answer?
1. Little or no difference between age groups (even bars, no slope):
Is the E-language (linguistic community) stable or do we have a
change that affects all age groups equally? (This would involve
“change” in the language of individuals (actual change or diffusion
(acquisition and use of a new grammar, according to Hale).)
2. Considerable difference between age groups (unev. bars, slope):
Is this a real change (in the wide sense, including diffusion-
induced changes in language use) and not age grading?
How, and to what extent, can the language of adults change and
what does that tell us about the nature of language, linguistic
knowledge, I-language ...?
(Keep in mind the difference between a trend study and a panel
study.)
Lund University,
September 6, 2012 23
Höskuldur Thráinsson
University of Iceland
Four Research Projects
1. Survey of phonological dialects in Iceland in the 1980s
(RÍN): Extensive overview of phonological dialects in
Iceland; some 2800 subjects; different age groups; all parts
of the country; partially a combination of a trend study and
a panel study since some of the subjects had been
“tested/interviewed” in the 1940s and the results were
being compared. (PIs Kristján Árnason and Höskuldur
Thráinsson, cf. Höskuldur and Kristján 1992, Kristján and
Höskuldur 2003).
2. Survey of a new syntactic construction in Icelandic (the
New Passive): An acceptability study (1999), using written
questionnaires. Some 1700 teenagers from all parts of the
country, plus 200 adults. (PIs Sigríður Sigurjónsdóttir and
Joan Maling, cf. Maling and Sigr. Sigurjónsdóttir 2002).
Lund University,
September 6, 2012 24
Höskuldur Thráinsson
University of Iceland
Four Projects ..., 2
3. Variation in Icelandic syntax (IceDiaSyn, 2005
2008), connected to similar projects in Scandinavia.
An extensive survey, mainly using written
questionnaires but also interviews, texts etc.
Approx. 3 x 700 speakers, different age groups,
different parts of the country. (PI Höskuldur
Thráinsson, cf. e.g Höskuldur et al. 2007, Höskuldur
forthcoming.). Included a detailed overview of the
New Passive in different age groups (cf. project 2).
4. Real time change in Icelandic phonology and syntax
(Raun, 20102012): Mostly a panel study
reinterviewing /-testing subjects that participated in
projects 1 and 2. (More information below.)
Lund University,
September 6, 2012 25
Höskuldur Thráinsson
University of Iceland
(Apparent) Evidence
for Linguistic Change in Apparent-Time
Fig. 3: Voiced sonorants before /p,t,k/ Fig. 4: ks-pronunciation
Correlation w. age: r = 0.385, p < 0,001 Correl. w. age: r = 0.422, p < 0.001
N>300 (area in Northeastern Icel., cf. below) N>2.800 (the whole country)
100 = the phonological feature never occurred in the subjects’ speech sample.
200 = the phonological feature always occurred where possible in the subjects’
speech sample
Lund University,
September 6, 2012 26
Höskuldur Thráinsson
University of Iceland
Apparent-Time, 2
Fig. 5: Topicalization in complem. clauses. Fig. 6: “Dative sickness” w. selected verbs.
Correl. w. age: r = 0.466, p < 0.001 Correl. w. age: r = 0.511, p < 0.001,
N > 700 N > 740
Data from judgment tasks. 1 = the construction rejected, 3 = the construction
accepted (cf. below note that the values 1 and 3 are sometimes be reversed,
but that will be made clear!)
Lund University,
September 6, 2012 27
Höskuldur Thráinsson
University of Iceland
Apparent-Time, 3
Fig. 7: Extended Progressive (to stat. vbs.). Fig. 8: The New Passive.
Correlation w. age: r = 0.442, p < 0.001. Correl. w. age: r = 0.642, p < 0.001.
N > 700 N > 720
Judgment tasks (cf. below. Data from the variation study (Project
3 on slide 25).
Lund University,
September 6, 2012 28
Höskuldur Thráinsson
University of Iceland
Apparent-Time, 4
Could figs. 34 represent age-grading rather than linguistic change
in apparent-time? What would that mean?
Fig. 3: Voiced sonorants before /p,t,k/ (e.g. mjólk ‘milk’) would then be
something that speakers gradually develop.
Fig. 4: ks-pronunciation (e.g. vaxa ‘grow’) would be a “habit” that speakers
“outgrow” (cf. Höskuldur and Kristján 1992, Kristján and Hösk. 2003).
Fig. 5: Topicalization in embedded clauses (“... that this song could he not
sing”) would be something that speakers gradually develop (cf.
Ásgrímur Angantýsson 2011 on emb. Top.).
Figs. 68: “Dative Sickness” (‘me (D) needs money’), the Extended Progressive
(“I’m not understanding this”) and the New Passive (“There was
pushed me”) would be “habits” that speakers “outgrow” (see e.g.
Höskuldur 2007, forthcoming; Maling and Sigr. Sigurjónsdóttir 2002;
Is this likely? Is this impossible? How can we study this?
Lund University,
September 6, 2012 29
Höskuldur Thráinsson
University of Iceland
Evidence for Linguistic Change in Real-Time
More information on the Real-Time Project (Project 4):
In the phonological part we are:
1. Reinterviewing some 700 speakers who participated in a
phonological project in the 1980s (Project 1 above).
Approx. 200 of these also participated in an earlier study
in the 1940s (and had “stayed at home”).
2. Interviewing some 100 speakers who participated in the
1940s-study and moved to “the city” (Reykjavík) but were
not included in RÍN (Project 1).
In the syntactic part we are:
1. Retesting some 200 speakers who participated in the
original study of the New Passive some 12 years ago
(Project 2 above).
2. Doing a separate “reliability check” of the questionnaire
method (redoing one of the IceDiaSyn tests, using the
same informants some 5 years later).
Lund University,
September 6, 2012 30
Höskuldur Thráinsson
University of Iceland
Real-Time Phonology, 1
Methodology of the phonological part:
1. Interviews centering around selected pictures to elicit
particular words (speech sounds, clusters ...):
2. Reading of short passages.
Purpose of methodology: To maximize comparability of the data.
Lund University,
September 6, 2012 31
Höskuldur Thráinsson
University of Iceland
Real-Time Phonol., 2
First example: Two North-Eastern phonological variants:
Hard pronunciation”: Aspirated (vs. unaspirated, voiceless)
/p,t,k/ after long vowels: api ‘ape’, tur ‘boat’, kaka ‘cake’.
Voiced pronunciation”: Voiced (vs. voiceless) /l,m,n/ before
/p,t,k/: mjólk ‘milk’, lampi ‘lamp’, banki ‘bank’ (and /ð/ befor
/k/: maðkur ‘worm’).
Distribution of these variants in the 1980s (RÍN, Project 1)
Hard stops Voiced sonorants
Lund University,
September 6, 2012 32
Höskuldur Thráinsson
University of Iceland
Real Time Phonol., 3
2nd example: Two Southern/South-Eastern phonol. variants:
Monophthongal pronunciation”: Monophthongs (vs.
diphthongs) before /gi/: magi ‘stomach’, bogi ‘bow’, lögin ‘the
laws’
hv-pronunciation”: hv- ([xv]) word-initially (as opposed to
[khv]): hvalur ‘whale’, hvað ‘what’
Distribution of these variants in the 1980s (RÍN, Project 1):
Monophthongal pron. hv-pronunciation Mon. & hv-: Age grps
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 33
Einhljóðaframburður og hv -framburður
í Vestur-Skaftafellssýslu
100
110
120
130
140
150
160
170
180
190
200
12 til 20 ára 21 til 45 ára 46 til 55 ára 56 til 70 ára 71 árs og e.
Einhlj.frb.
Hv-frb.
Real-Time Phonol., 4
Question:
Are the regional phonological variants just illustrated
retreating because the new generations do not
acquire them or are the speakers in the region
gradually giving them up (cf. the main question:
“How do languages change?”).
Lund University,
September 6, 2012 34
Höskuldur Thráinsson
University of Iceland
Real-Time Phonol., 5
Schematic comparison of three phonol. studies:
Subjects studied in BG (1940+), RÍN (1980+) and RAUN (2010+).
Arrows and colors showing repeated interviews with the same
subjects.
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 35
Real-Time Phonol., 6
Comments on the methodology:
The elicitation techniques were very similar and the
social conditions were virtually identical in the three
phonological studies. Typically the speakers had not
moved and were interviewed at home. Hence the
changes may not be “reversible(in Hale’s sense).
(One group is an exception to this, the “moved”
subjects discussed below.)
The features are regional, although the “hard
pronunciation” and the hv-pronunciation are
somewhat more widespread than the others and
perhaps more highly regarded by some (although not a
part of “the standard Reykjavík dialect”).
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 36
Real-Time Phonol., 7
Fig. 9: The development of three regional features (means) in the
speech of 15 speakers, tested three times over the period of 70 years
(cf. Margrét Lára Höskuldsdóttir 2012).
Lund University,
September 6, 2012 37
Höskuldur Thráinsson
University of Iceland
Real-Time Phonol., 8
An additional question:
Are younger speakers more likely than older ones to “change
their pronunciation” (cf. Katrín María Víðisdóttir 2011)?
Fig. 10: The development of voiced pronunciation from the
1980s (RÍN) to 2010 (Raun). Older group (10 speakers who were
over 50 in RÍN). Means 150 vs. 140 (difference 10).
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 38
Real-Time Phonol., 9
Change in voiced pronunciation among younger speakers:
Fig. 10: The development of voiced pronunciation from the
1980s (RÍN) to 2010 (Raun). Younger group (10 speakers who
were under 50 in RÍN). Means 160 vs. 140 (difference 20).
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 39
Real-Time Phonol., 10
A final question on the phonology:
Are speakers who moved out of the dialectal area more
likely to “change their pronunciation” than those who stayed
behind (cf. Margrét Lára Höskuldsdóttir 2012)?
Fig.11: Development of three regional features from the 1940s to
2011. A group of 36 speakers who moved vs. 41 who did not.
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 40
Real-Time Phonol., 11
More on the North-Eastern features
(data from 37 subjects from Eyjafjörður):
Hard stops in RÍN vs. RAUN Voiced son. in RÍN vs. RAUN
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 41
Real-Time phonol., 12
Some information on the Southern/South-Eastern
features (22 speakers from V-Skaftafellssýsla):
Monophthongs in RÍN vs. RAUN hv-pron. in RÍN vs. RAUN
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 42
Interim conclusion
Some of the phonological regional variants are not
only losing ground because new generations do not
adopt them/acquire them but also because adult
speakers are gradually giving them up.
This does not hold to equal extent for all variants.
The reasons for these differences may be structural
rather than social (sociolinguistic).
An interesting case: The hv-pronunciation (can be
discussed later if there is time and/or inclination).
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 43
Real-Time Syntax, 1
Methodology of the syntactic part: Most importantly a
judgment task, using a written questionnaire. Illustration (the
Nw Passive, English glosses were not included!)
Lund University,
September 6, 2012 44
Höskuldur Thráinsson
University of Iceland
Real-Time Syntax, 2
An important methodological question:
How reliable are the speakers’ judgments in
written questionnaires like this?
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 45
Real-Time Syntax, 3
Some indications that the questionnaire-methodology
works (“the proof is in the pudding”, cf. Höskuldur
Thráinsson 2012):
The variation is systematic (differences between clause
types, age groups and (in a few cases) regions, etc.) and not
random.
All generations seem reliable (e.g., it’s not the case that the
youngest generation “accepts everything”).
The subjects answer honestly in general and don’t seem
worried by any kind of prescriptivism or the like (there is very
little awareness of most of the variants investigated anyway).
Comparison of different tasks confirms reliability of
judgments.
Comparison with corpora confirms reliabilty of judgments (cf.
Ásta Svavarsdóttir 2012).
Comparison with interviews confirms reliability of judgments
(cf. Höskuldur Thráinsson et al. eds. 2012)
...
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 46
Real-Time Syntax, 4
Testing the reliability of judgments:
A new study, which is a repetition of “IceDiaSyn 2”,
testing the same speakers some 5 years later, using
the exact same questionnaire (testing e.g. Dative
Substitution, Object Case, selection of
Indicative/Subjunctive in embedded clauses,
“oblique null PRO” and verb agreement with
partitive genitives ...)
So far I only have results from 6 speakers.
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 47
Real-Time Syntax, 5
Examples of sentences included in the “Reliability Study”:
(1)a. Margir fara á árshátíð í febrúar.
many go to annual-celebration in February
Okkur Þorvaldi langar að fara á þorrablót. DS (5)
me and Thorvald(D) want to go to “þorrablót”
b. Tölvutæknin hefur valdið byltingu.
the computer technology has caused a revolution
Fólk þarf sjaldan að faxa eyðublöðum lengur.
people need rarely to fax forms(D) longer Dat(vs. Acc.) obj. (5)
c. Gummi er búinn að vera veikur í tvo daga.
Gummi has been sick for two days Ind. vs.
Hann ætlar samt í ferðalagið þótt hann er enn mjög slappur. sbjnct. (6)
d. Starfsmennirnir lærðu slökunaræfingar á námskeiðinu.
the employees learned relaxation-exercises in the course
Nú getur Ósk hlakkað til að líða betur á morgun. Obl. PRO (4)
now can Ósk look forward to PRO(Dat) feel better tomorrow
e. Skólaballið stóð langt fram á nótt.
the school-dance lasted long into night
Fjöldi nemenda vöknuðu of seint morguninn eftir. Part.Gen agr. (4)
number (of) students(Gen) woke up too late next morning
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 48
Real-Time Syntax, 6
Preliminary results from the reliability study:
A total of 207 judgments by these 6 speakers.
74% (154) of these were exactly the same 5 years later
In 14% (28) of the instances there were “minimal
differences”, i.e. the speakers had selected “yes” (‘natural
sentence’) in one study and “?” (‘questionable’) in the other
or “?” in one study and “no” (‘unnatural sentence’) in the
other
In 12% the speakers had reversed their judgments (“yes”
instead of “no” or vice versa). (Note that some of these
answers might reflect a “real change”, cf. below.)
So there is undoubtedly some “noise” in data of this kind but
probably less than most people would have expected.
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 49
Real-Time syntax, 7
The main study has to do with the “New Passive” (cf. e.g.
Maling and Sigr. Sigurjónsdóttir 2002, Halldór Ármann
Sigurðsson 2011, Einar Freyr Sigurðsson 2012):
(1) Active: Einhver skammaði strákinn.
somebody scolded the-boy(A)
(2) a. Canonical passive: Strákurinn var skammaður.
the-boy(N.sg.m.) was scolded(sg.m.)
b. New passive: Það var skammað strákinn.
there was scolded(sg.n) the-boy(A.sg.m.)
Icelandic has an expletive passive:
(3) a. Einhver strákur var skammaður.
some boy(N.sg.m.) was scolded(sg.m.)
b. Það var skammaður einhver strákur.
there was scolded(sg.m.) some boy(N.sg.m.)
The (postposed) subject in an expletive passive has to be indefinite, as
normally in expletive constructions. Note also the case.
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 50
Real-Time syntax, 8
The New Passive was only noticed quite recently and it was
typically rejected by the adults in the original study by Maling &
Sigr. Sigurjónsdóttir 1999 (M&S, Project 2 on slide 25 above). We
got comparable results in IceDiaSyn (Project 3), cf. below:
Fig. 12: Preliminary comparison of the results of IceDiaSyn 2005
(first 4 columns) and M&S 1999.
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 51
Real-Time syntax, 9
Two points on the comparison of IceDiaSyn and M&S:
This construction does not seem to spread to older
generations (no diffusion?)
One might have expected greater acceptance by the 2025
year olds in IceDiaSyn since this is partially the same
generation as the one tested by M&S. Do speakers outgrow
this?
So are doing a Real-Time study (the syntactic part of Project 4):
142 speakers who “tested” by Maling and Sigr. Sigurjóns-
dóttir 1999 and “retested” 2010–2012. They come from
different parts of the country.
6 typical “New Passive” examples tested in exactly the same
way (except that context sentences were included this time).
In the part reported on here, the subjects could only answer
“yes” and “no” (no “?”).
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 52
Real-Time syntax, 10
The NewPass examples reported on here:
(1)a. Nokkrir strákar svindluðu á prófinu.
some boys cheated on the exam
Það var rekið Ólaf úr skólanum.
there was expelled Olaf(Acc) from school
b. Uppþvottavélin var biluð í gær.
the dish-washer was broken yesterday
Það var beðið mig að vaska upp.
there was asked me(Acc) to wash up
c. Leigjendurnir skildu eftir fullt af dóti í íbúðinni.
the tenants left lots of stuff in the apartment
Það var fleygt draslinu á haugana.
there was thrown the trash(Acc) on the garbage-heap
d. Ég gat ekki farið í bíó í gærkvöldi.
I could not go to cinema last night
Það var sagt mér að taka til.
there was told me(Dat) to clean up
e. Það var skilið hana eftir heima.
there was left her(Acc) at home
f. Í gær var tekið lyklana af honum.
yesterday was taken the keys(Acc) from him
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 53
Real-Time syntax, 11
Comparison of the judgments 1999 and 20102012:
Mean Scores (1 = accepts all, 3 = rejects all):
1999: 1,93
20102012: 2,57
71% (101) of the speakers judge the example
sentences less positively now than 1999
27% (38) of the speakers judge the sentences the
same way (on the average) now as 1999.
2% (3) speakers are more positive towards the New
Passive examples now than in 1999.
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 54
Real-Time syntax, 12
Although most of the speakers accept the New
Passive examples less readily 10 years later (when
they are approx. 25 years old vs. 15 years old), they
typically have not completely “outgrown the habit”:
There is a strong correlation between the judgments
1999 and now (r = 0,549, p < 0,001)
There is a statistically significant correlation with
education now, i.e. the more educated speakers are
less likely to accept the New Passive now (r = 0.356,
p< 0,001)
(r > 0,5 is strong correlation, r = 0,30,5 is medium correlation
and r = 0,10,3 is weak correlation, cf. Field 2005:32)
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 55
Concluding remarks
Speakers can and do change their “language” in
some sense.
This applies both to (phonological) production and
(syntactic) judgments.
This sheds some light on the nature of linguistic
change, although it is not and easy task to determine
how “deep” this phenomenon is nor to fit this into
a model of the speakers’ linguistic knowledge.
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 56
References
Andersen, Henning. 1973. Abductive and Deductive Change. Language 49:765
793.
Anderson, Stephen R., and David Lightfoot. 2002. The Language Organ:
Linguistics as a Cognitive Psychology. Cambridge University Press, Cambridge.
Ásgrímur Angantýsson. 2011. The Syntax of Embedded Clauses in Icelandic and
Related Languages. PhD dissertation, University of Iceland, Reykjavík.
Chomsky, Noam. 1965. Aspects of the Theory of Syntax. MIT Press, Cambridge,
Mass.
Chomsky, Noam. 1986. Knowledge of Language. Its Nature, Origin and Use.
Praeger, New York.
Einar Freyr Sigurðsson. 2012. Germynd en samt þolmynd. Um nýju þolmyndina í
íslensku. MA thesis, University of Iceland.
Grace, George. Speaking of Language Change. University of Hawaii Working
Papers in Linguistics 3:101116.
Hale, Mark. 2007. Historical Linguistics: Theory and Method. Blackwell, Oxford.
Halldór Ármann Sigurðsson. 2011. On the New Passive. Syntax 14(2):148178.
Halle, Morris. 1962. Phonology in Generative Grammar. Word 18:5472.
Hreinn Benediktsson. 2002. The Vowel System of Icelandic: A Survey of Its
History. In: Guðrún Þórhallsdóttir, Höskuldur Þráinsson, Jón G. Friðjónsson
and Kjartan Ottosson (eds.): Linguistic Studies, Historical and Comparative by
Hreinn Benediktsson, pp. 5073. Linguistic Institute, University of Iceland,
Reykjavík. [Originally published in Word 15:282312, 1959.]
Lund University,
September 6, 2012 57
Höskuldur Thráinsson
University of Iceland
References, 2
Höskuldur Thráinsson. 1980. Sonorant Devoicing at Lake Mývatn: A Change in
Progress. In: Even Hovdhaugen (ed.): The Nordic Languages and Modern
Linguistics [4], pp. 355364. Universitetsforlaget, Oslo.
Höskuldur Thráinsson. 2007. The Syntax of Icelandic. Cambridge University
Press, Cambridge.
Höskuldur Thráinsson. 2012. Ideal Speakers and Other Speakers The case of
dative and some other cases. In Beatriz Fernández and Ricardo Etxepare
(eds.): Variation in datives: A Micro-Comparative Perspective, pp. 161-
188. Oxford Studies in Comparative Syntax, Oxford University Press,
Oxford.
Höskuldur Thráinsson, Ásgrímur Angantýsson, Ásta Svavarsdóttir, Thórhallur
Eythórsson and Jóhannes Gísli Jónsson. 2007. The Icelandic (Pilot) Project
in ScanDiaSyn. Nordlyd 34,1: 87-124. Special issue on Scandinavian Dialect
Syntax 2005.
Höskuldur Thráinsson and Kristján Árnason. 1992. Phonological Variation in
20th Century Icelandic. Íslenskt mál 14:89128.
Isac, Daniela, and Charles Reiss. 2008. I-Language. An Introduction to
Linguistics as Cognitive Science. Oxford University Press, Oxford.
Lund University,
September 6, 2012 58
Höskuldur Thráinsson
University of Iceland
References, 3
Katrín María Víðisdóttir. 2011. Þá og nú. Um einstaklingsþróun á mállýskum á
Norðurlandi frá níunda áratugnum þar til nú. [‘Then and Now. On the
development of regional Northern features in the speech of individuals
from the 1980s until now.’] BA-thesis, University of Iceland, Reykjavík.
Kristján Árnason and Höskuldur Thráinsson. 2003. Fonologiske dialekttræk på
Island. Generationer og geografiske områder. In Gunnstein Akselberg,
Anne Marit Bødal and Helge Sandøy (eds.): Nordisk dialektologi, pp. 151
196. Novus, Oslo.
Labov, William. 1963. The Social Motivation of a Sound Change. Word 19:273-
309. [Cf. Chapter 1 of Labov 1972.]
Labov, William. 1966. The Social Stratification of English in New York City.
Center for Applied Linguistics, Washington D.C.
Labov, William. 1972. Sociolinguistic Patterns. University of Pennsylvania
Press, Philadelphia.
Labov, William. 2001. Principles of linguistic change. Volume I: Internal
Factors. 2nd edition [with a new preface; fyrst ed. 1994]. Blackwell, Oxford.
Lenneberg, Eric H. 1967. The Biological Foundations of Language. Wiley, New
York.
Lightfoot, David W. 1979. Principles of Diachronic Syntax. Cambridge
University Press, Cambridge.
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 59
References, 4
Lightfoot, David W. 1999. The Development of Language: Acquisition, Change, and
Evolution. Blackwell, Oxford.
Lightfoot, David W. 2006. How New Languages Emerge. Cambridge University
Press, Cambridge.
Maling, Joan, and Sigríður Sigurjónsdóttir. 2002. The ‘New Impersonal’ Construction
in Icelandic. Journal of Comparative Germanic Linguistics 5:97142.
Margrét Guðmundsdóttir. 2008. Málbreytingar í ljósi málkunnáttufræði. [´Linguistic
change in the light of generative grammar.’] Íslenskt mál 30:7‒52.
Margrét Lára Höskuldsdóttir. 2012. Ertu nokkuð búin(n) að [tha:pha] þér?
Framburður Norðlendinga á Norðausturlandi og í Reykjavík á árunum 1940
2011. [‘Have you lost it? The pronunciation of Northerners in North-East Iceland
and Reykjavík 1940–2011.’] BA-thesis, University of Iceland, Reykjavík.
Sankoff, Gillian, and Hélène Blondeau. 2007. Language Change across the Lifespan:
/r/ in Montreal French. Language 83:560588.
Weinreich, Uriel, William Labov and Marvin Herzog. 1968. Empirical Foundations
for a Theory of Language Change. Winfred P. Lehmann and Yakov Malkiel (eds.):
Directions for Historical Linguistics, bls. 95-188. University of Texas Press,
Austin.
Lund University,
September 6, 2012 Höskuldur Thráinsson
University of Iceland 60
ResearchGate has not been able to resolve any citations for this publication.
Book
The Language Organ, first published in 2002, treats human language as the manifestation of a faculty of the mind, a mental organ whose nature is determined by human biology and whose functional properties should be explored just as physiology explores the functional properties of physical organs. It surveys the nature of the language faculty in its various aspects: the systems of sounds, words, and syntax, the development of language in the child and historically, and what is known about its relation to the brain. It discusses the kinds of work that can be carried out in these areas that will contribute to an understanding of the human language organ. This book will appeal to students and researchers in linguistics, and is written to be accessible to colleagues in other disciplines dealing with language as well as to readers with an interest in general science and the nature of the human mind.
Article
Understanding the relationship between language change and variation has progressed considerably over the last several decades, but less is known about how speakers at different life stages deal with ongoing change in their speech communities. Longitudinal studies of individuals and groups reveal three trajectory types postadolescence: stability (the most common), adopting (to some degree) a change led by younger people (the next most common trajectory), or swimming against the community current by reverting to an older pattern in later life (the least common trajectory). Declining plasticity over the life course places limits on possible trajectories, which are also subject to social and cultural influences. This article reviews relevant studies from historical linguistics as well as panel studies on African American English and dialect contact, proposing that future progress will be made by interdisciplinary research combining psycholinguistic and sociolinguistic perspectives. Lifespan trajectories in situations of community stability are also discussed.
Book
The coming of language occurs at about the same age in every healthy child throughout the world, strongly supporting the concept that genetically determined processes of maturation, rather than environmental influences, underlie capacity for speech and verbal understanding. Dr. Lenneberg points out the implications of this concept for the therapeutic and educational approach to children with hearing or speech deficits.
Article
New languages are constantly emerging, as existing languages diverge into different forms. To explain this fascinating process, we need to understand how languages change and how they emerge in children. In this pioneering study, David Lightfoot explains how languages come into being, arguing that children are the driving force. He explores how new systems arise, how they are acquired by children, and how adults and children play different, complementary roles in language change. Lightfoot makes an important distinction between ‘external language’ (language as it exists in the world), and ‘internal language’ (language as represented in an individual's brain). By examining the interplay between the two, he shows how children are ‘cue-based’ learners, who scan their external linguistic environment for new structures, making sense of the world outside in order to build their internal language. Engaging and original, this book offers an interesting account of language acquisition, variation and change.