Content uploaded by Daniil M. Ozernyi
Author content
All content in this area was uploaded by Daniil M. Ozernyi on Mar 29, 2022
Content may be subject to copyright.
Gist Current models Epistemology Conclusive remarks References
On meronomy and dimensionality
of the models for multilingual
language acquisition
Daniil M. Ozernyi
Department of Linguistics,
Northwestern University
UNC-Chapel Hill Spring Colloqium
March 26, 2022
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 1 / 27
Gist Current models Epistemology Conclusive remarks References
Gist
The broad argument of this paper is as follows.
The broad meronomy thesis
Theorizing about the process of Lnacquisition requires
commitment to a particular view of language as a phenomenon.
The reasons for introducing such a thesis come principally from
epistemology and philosophy of science. However, exigence for
such a thesis comes principally from seeming lack of soundness of
some recent investigations into Lnacquisition.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 2 / 27
Gist Current models Epistemology Conclusive remarks References
Conjectures present in the field
1. Cumulative-Enhancement Model
(Flynn et al., 2004)
2. Micro-cue and its versions
(Westergaard, 2021)
3. Grammatical Mapping for L3 (Flynn
and Fern´andez-Berkes, n.d.)
4. The Scalpel Model (Slabakova, 2017)
5. Language of Community (Fallah
et al., 2016)
6. L1 Privilege (Hermas, 2010)
7. L2 Status Factor (Falk and Bardel,
2011 et seq.)
8. Typological Proximity Model
(Rothman, 2010)
9. Linguistic Proximity Model
(Mykhaylyk et al., 2015)
10. Interlanguage Transfer Hypothesis
(Leung, 2007)
11. FTFA for L3(Schwartz and Sprouse,
2021)
12. Phonological Permeability Hypothesis
(Amaro, 2017)
13. Similarity Convergence Hypothesis
(Brown and Chang, 2022)
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 3 / 27
Gist Current models Epistemology Conclusive remarks References
Generativism
Some of the models above have been cast as generative models
(e.g., CEM). Some of them are traditionally assumed as
generative by proxy, because of association of their authors with
the generative tradition broadly (e.g., Scalpel or Micro-cue).
The question now is this: can a model be somehow “moderatly
generative”, e.g., subscribe to some tenets of generativism while
not subscribe to the others? Can a model belong to a larger field
of language acquisition?
That is to say: can, for example, Language of Community be
somehow “partially generative” model? After all, the authors
draw syntactic trees, talk of headedness, and cite Kayne, 1994..?
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 4 / 27
Gist Current models Epistemology Conclusive remarks References
Generativism
Some of the models above have been cast as generative models
(e.g., CEM). Some of them are traditionally assumed as
generative by proxy, because of association of their authors with
the generative tradition broadly (e.g., Scalpel or Micro-cue).
The question now is this: can a model be somehow “moderatly
generative”, e.g., subscribe to some tenets of generativism while
not subscribe to the others? Can a model belong to a larger field
of language acquisition?
That is to say: can, for example, Language of Community be
somehow “partially generative” model? After all, the authors
draw syntactic trees, talk of headedness, and cite Kayne, 1994..?
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 4 / 27
Gist Current models Epistemology Conclusive remarks References
Generativism
Some of the models above have been cast as generative models
(e.g., CEM). Some of them are traditionally assumed as
generative by proxy, because of association of their authors with
the generative tradition broadly (e.g., Scalpel or Micro-cue).
The question now is this: can a model be somehow “moderatly
generative”, e.g., subscribe to some tenets of generativism while
not subscribe to the others? Can a model belong to a larger field
of language acquisition?
That is to say: can, for example, Language of Community be
somehow “partially generative” model? After all, the authors
draw syntactic trees, talk of headedness, and cite Kayne, 1994..?
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 4 / 27
Gist Current models Epistemology Conclusive remarks References
First, PoS
It is clear as a bell that hypotheses cannot be tested in isolation.
This much was not argued against since it was asserted by Duhem
(Duhem, 1909). Similar remarks appeared in Quine (cf. broadly
sect. 6 in Quine, 1951).
We shall take it as given that no experiment can test a hypothesis
in isolation. Then, however, we come to the question – if there is
a system of propositions (call the “beliefs”), how does it function?
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 5 / 27
Gist Current models Epistemology Conclusive remarks References
First, PoS
It is clear as a bell that hypotheses cannot be tested in isolation.
This much was not argued against since it was asserted by Duhem
(Duhem, 1909). Similar remarks appeared in Quine (cf. broadly
sect. 6 in Quine, 1951).
We shall take it as given that no experiment can test a hypothesis
in isolation. Then, however, we come to the question – if there is
a system of propositions (call the “beliefs”), how does it function?
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 5 / 27
Gist Current models Epistemology Conclusive remarks References
Foundationalism
One view is that the set of beliefs is derived from a foundational
one. I.e., a set of beliefs b1...bnalways depends on some belief b0.
This is the foundationalist view, which we’ll take in the weaker
form of doxastic basicality (as opposed to epistemic basicality).
Doxastic Basicality (sec. Steup and Neta, 2020)
S’s justified belief that p is basic if and only if S’s belief that p is
justified without owing its justification to any of S’s other beliefs.
(For further background, cf., e.g., Chisholm, 1966, or an overview
in Steup, 1996, &c. – on which we shall draw freely forth.)
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 6 / 27
Gist Current models Epistemology Conclusive remarks References
Foundationalism
One view is that the set of beliefs is derived from a foundational
one. I.e., a set of beliefs b1...bnalways depends on some belief b0.
This is the foundationalist view, which we’ll take in the weaker
form of doxastic basicality (as opposed to epistemic basicality).
Doxastic Basicality (sec. Steup and Neta, 2020)
S’s justified belief that p is basic if and only if S’s belief that p is
justified without owing its justification to any of S’s other beliefs.
(For further background, cf., e.g., Chisholm, 1966, or an overview
in Steup, 1996, &c. – on which we shall draw freely forth.)
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 6 / 27
Gist Current models Epistemology Conclusive remarks References
A case study
Take Language of Community. The authors write that “it seems
that in the initial stages of L3 acquisition, CLI originates from the
dominant language of communication, irrespective of whether it is
the L1 or L2”, making explicit that “dominant language of
communication” plays a crucial role (place of origin) for
decision-making of key aspects of the acquisition process
(CLIs).
Let us assume that processes involving CLIs are such that take
place within I-language1.We can then derive a contradiction.
1We arrive at this by reductio that if it isn’t, then Lnacquisition is a
process independent of I-language, and cannot be of generative salience.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 7 / 27
Gist Current models Epistemology Conclusive remarks References
A case study
Take Language of Community. The authors write that “it seems
that in the initial stages of L3 acquisition, CLI originates from the
dominant language of communication, irrespective of whether it is
the L1 or L2”, making explicit that “dominant language of
communication” plays a crucial role (place of origin) for
decision-making of key aspects of the acquisition process
(CLIs).
Let us assume that processes involving CLIs are such that take
place within I-language1.We can then derive a contradiction.
1We arrive at this by reductio that if it isn’t, then Lnacquisition is a
process independent of I-language, and cannot be of generative salience.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 7 / 27
Gist Current models Epistemology Conclusive remarks References
Contradiction for LoC 1
⊥
...
a0...
...
... b0
Let foundational beliefs be:
ao: Communication can play
some salient role for I-language.
b0: E-Language is peripheral,
marginal for generative inquiry
into acquisition.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 8 / 27
Gist Current models Epistemology Conclusive remarks References
Elaboration
You can either ascribe:
to the view that (communication ∈) E-language is crucial for
acquisition,
or to the generative view of language as a phenomena (which
has rejected the role E-language).
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 9 / 27
Gist Current models Epistemology Conclusive remarks References
Cont’d
Implying influence of E-language in I-language (as LoC does) is
incoherent. It might appear that one might very well take
generative trees and Kayne’s antisymmetry (&c.) and mix it in
with the belief in salience of E-language, but this fails to account
that antisymmetry arose from some other considerations,
ultimately boiling down to broad approach to language articulated
in LSLT (Chomsky, 1955) and Aspects (Chomsky, 1965).
This approach, in turn, is incompatible with salience of
E-language. It fails any scrutiny to assume connection between
PSRs, transformations, and other vital parts of generativism with
communication. Then, by extension, contradiction on an earlier
slide holds. Assumptions of generativism ought never be taken
out of context.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 10 / 27
Gist Current models Epistemology Conclusive remarks References
Coherentism
However, foundationalism is not the only view in justification and
belief structure. Let’s consider an (arguably viable) alternative to
foundationalism: coherentism. (Most straightforwardly taken as
rejection of doxastic basicality, though cf. dependence
coherentism.)
Doxastic Coherentism (sec. Steup and Neta, 2020)
Every justified belief receives its justification from other beliefs in
its epistemic neighborhood.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 11 / 27
Gist Current models Epistemology Conclusive remarks References
Fair warning!
Not representing a review of the TPM here
I do not take TPM’s last renditions for the discussion in
subsequent slides. While I do think that the broad points about
the TPM that are made below are veridical, I’d like to
acknowledge that the purpose of discussion below is aimed at
showing that doxastic coherentism does not rescue a model
from contradictions. This is not a comprehehsive review of
TPM.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 13 / 27
Gist Current models Epistemology Conclusive remarks References
Another case study
Take, say, the TPM. Then define (i) b1: “Typology plays a salient
role in acquisition”, (ii) b2: “There is initial state for L3
acquisition sensu initial state for L1”, (iii) b3: “We assume
minimalist framework of Chomsky, 1995”2.
But again, this yields contradictions, for valuations
v(b1∧b3) =⊥
v(b2∧b3) =⊥(assume inheriting Chomsky, 1986 in
Chomsky, 1995)
These ought to be explicated further:
2“I appeal to current proposals of the composition of the language faculty and the
feature-based linguistic computational system as articulated in the Minimalist Program”
(Rothman, 2015, p.181).
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 14 / 27
Gist Current models Epistemology Conclusive remarks References
Another case study
Take, say, the TPM. Then define (i) b1: “Typology plays a salient
role in acquisition”, (ii) b2: “There is initial state for L3
acquisition sensu initial state for L1”, (iii) b3: “We assume
minimalist framework of Chomsky, 1995”2.
But again, this yields contradictions, for valuations
v(b1∧b3) =⊥
v(b2∧b3) =⊥(assume inheriting Chomsky, 1986 in
Chomsky, 1995)
These ought to be explicated further:
2“I appeal to current proposals of the composition of the language faculty and the
feature-based linguistic computational system as articulated in the Minimalist Program”
(Rothman, 2015, p.181).
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 14 / 27
Gist Current models Epistemology Conclusive remarks References
(i) v(b1∧b3) =⊥
(i) b1: “Typology plays a salient role in acquisition”,
(iii) b3: “We assume minimalist framework of Chomsky, 1995”
The minimalist framework assumes such an architecture of
language (“composition of the language faculty and the
feature-based linguistic computational system”) that typology is
not well-defined in such a system. Typology, in fact – if taken to
be a difference in feature-valuations3– reduces one quickly and
easily into Sorites paradox. If one assumes the minimalist
framework, then micro-variation on a single feature-valuation
yields “a different language”. Then, most speakers of what we call
“English” speak different languages.
3Also, are features really an account of crosslingustic variation?
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 15 / 27
Gist Current models Epistemology Conclusive remarks References
Cont’d
This is not as important for cross-linguistic work, where the only
thing that matters is a comparison between two different systems
(no matter whether “languages”, or anything else – e.g., dialects,
micro-variations in an otherwise coherent linguistic community).
For acquisition work, no such approach can be adopted.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 16 / 27
Gist Current models Epistemology Conclusive remarks References
(ii) v(b2∧b3) =⊥
(ii) b2: “There is initial state for L3acquisition sensu initial state
for L1”,
(iii) b3: “We assume minimalist framework of Chomsky, 1995”.
This is explored at length in Ozernyi (20224), but the gist is that
for Chomsky (in KoL), “[UG] is taken to be the set of properties,
conditions, or whatever that constitute the ‘initial state’ of the
language learner, hence the basis on which knowledge of language
develops”.
Once valuations (features, parameters, something else) are set (at
least one of them) then initial state is gone, over, no more.
Further, per linguistic universals – once some universal u1is set of
one language, it is set for all others.
4Upcoming talk at UC Berkeley on Apr 9, 2022.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 17 / 27
Gist Current models Epistemology Conclusive remarks References
What of epistemology?
So, both coherentism and foundationalism allows us to see
conjunctions of beliefs which result in contradictions. These
contradictions are not easily detected, particularly in the cases
when a number of layers between contradictory beliefs of great.
However, we ought not forget that no matter how far from each
other contradictory beliefs are, they still yield a contradiction.
¬B∧...ϕ... ∧...ψ... ∧B⊢⊥ provided B, ¬B /∈ϕ, ψ
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 18 / 27
Gist Current models Epistemology Conclusive remarks References
NB
We could go through and point out how certain elements of
conjectures from slide 3 (most prominently from 2, 4, 6-13)
contradict the generative framework – and are best situated
explicitly outside such framework.
This clear situation will both relieve the models which do not
claim their generativity from unnecessary objections posed by
contradictions with the generative framework – and also confine
the focus of researcher working on generative approaches to
acquisition to those few models that are within the scope of
requisite approach. This would save time, resources, and heated
debates.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 19 / 27
Gist Current models Epistemology Conclusive remarks References
Conclusive remarks
As such, it is necessary to be very wary of the assumptions one is
making with regard to language and how to investigate it,
particularly for the purposes of acquisition inquiry.
We repeat, then, the slightly edited broad meronomy thesis
sketched at the beginning of the presentation:
The strong meronomy thesis
Theorizing about the process of Lnacquisition requires full
commitment to a particular view of language as a phenomenon.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 20 / 27
Gist Current models Epistemology Conclusive remarks References
Where else from?
We even content that contradictions can be derived within the
generative framework itself, and quite straightforwardly so.
E.g., by mixing beliefs from LFG/HPSG/GPSG and from
Minimalism. Further, they can be derived within Minimalism.
E.g., by both adopting Chomskian “economy” and Citko’s
multidominance/paralell merge5.
The big message is this: one cannot be too careful in clarifying
one’s assumptions when theorizing about acquisition process (and
checking them for contradictions).
5Note that the two are different.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 21 / 27
Gist Current models Epistemology Conclusive remarks References
References I
Amaro, J. C. (2017). Testing the Phonological Permeability Hypothesis: L3 phonological effects on L1
versus L2 systems. International Journal of Bilingualism,21(6), 698–717.
https://doi.org/10.1177/1367006916637287
Brown, M. M., & Chang. (2022). Regressive Cross-Linguistic Influence in Multilingual Speech Rhythm:
The Primacy of Typological Similarity.
Chisholm, R. M. (1966). Theory of Knowledge.Englewoo d Cliffs, NJ, USA: Englewood Cliffs, N.J.,
Prentice-Hall.
Chomsky, N. (1955). The Logical Structure of Linguistic Theory.Manuscript.
http://alpha-leonis.lids.mit.edu/wordpress/?page id=466
Chomsky, N. (1965). Aspects of the Theory of Syntax.MIT Press.
Chomsky, N. (1986). Knowledge of Language: Its Nature, Origin, and Use.
Chomsky, N. (1995). The Minimalist Program.The MIT Press. Retrieved June 1, 2021, from
https://muse.jhu.edu/book/36980
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 22 / 27
Gist Current models Epistemology Conclusive remarks References
References II
Duhem, P. M. M. (1909). The Aim and Structure of Physical Theory.Princeton University Press. Retrieved
March 26, 2022, from https://muse.jhu.edu/b ook/85196/
Falk, Y., & Bardel, C. (2011). Object pronouns in German L3 syntax: Evidence for the L2 status factor
[Publisher: SAGE Publications Ltd]. Second Language Research,27(1), 59–82.
https://doi.org/10.1177/0267658310386647
Fallah, N., Jabbari, A. A., & Fazilatfar, A. M. (2016). Source(s) of syntactic cross-linguistic influence
(CLI): The case of L3 acquisition of English possessives by Mazandarani–Persian bilinguals
[Publisher: SAGE Publications Ltd]. Second Language Research,32(2), 225–245.
https://doi.org/10.1177/0267658315618009
Flynn, S., & Fern´andez-Berkes, ´
E. (n.d.). Grammatical Mapping in L3 Acquisition: A Theory of
Development. In M. M. Brown, ´
E. Fern´andez-Berkes, & S. Flynn (Eds.), L3 Development After
the Initial State. John Benjamins Publishing Company.
Flynn, S., Foley, C., & Vinnitskaya, I. (2004). The cumulative-enhancement model for language
acquisition: Comparing adults’ and children’s patterns of development in first, second and
third language acquisition of relative clauses [Publisher: Taylor & Francis]. International
Journal of Multilingualism,1(1), 3–16.
https://www.tandfonline.com/doi/abs/10.1080/14790710408668175
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 23 / 27
Gist Current models Epistemology Conclusive remarks References
References III
Hermas, A. (2010). Language acquisition as computational resetting: Verb movement in L3 initial state.
International Journal of Multilingualism,7(4), 343–362.
https://doi.org/10.1080/14790718.2010.487941
Kayne, R. S. (1994). The Antisymmetry of Syntax (S. J. Keyser, Ed.). MIT Press.
Leung, Y.-k. I. (2007). Third language acquisition: Why it is interesting to generative linguists. Second
Language Research,23(1), 95–114. https://hal.archives-ouvertes.fr/hal-
00570723/file/PEER stage2 10.1177%252F0267658307071604.pdf
Mykhaylyk, R., Mitrofanova, N., Rodina, Y., & Westergaard, M. (2015). The Linguistic Proximity Model:
The Case of Verb-Second revisited. Proceedings of the 39st annual Boston University Conference
on Language Development, 337–349.
Quine, W. V. (1951). Main Trends in Recent Philosophy: Two Dogmas of Empiricism [Publisher: [Duke
University Press, Philosophical Review]]. The Philosophical Review,60(1), 20–43.
https://doi.org/10.2307/2181906
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 24 / 27
Gist Current models Epistemology Conclusive remarks References
References IV
Rothman, J. (2010). On the typological economy of syntactic transfer: Word order and relative clause
high/low attachment preference in L3 Brazilian Portuguese [Publisher: De Gruyter Mouton
Section: International Review of Applied Linguistics in Language Teaching]. 48(2-3), 245–273.
https://doi.org/10.1515/iral.2010.011
Rothman, J. (2015). Linguistic and cognitive motivations for the Typological Primacy Mo del (TPM) of
third language (L3) transfer: Timing of acquisition and proficiency considered [Publisher:
Cambridge University Press]. Bilingualism: language and cognition,18(2), 179–190.
https://www.researchgate.net/profile/Jason Rothman2/publication/
264295396 Linguistic and cognitive motivations for the Typ ological Primacy Model TPM of
third language L3 transfer Timing of acquisition and proficiency considered/links/
53d7b01f0cf2a19eee7fcc40/Linguistic-and- cognitive-motivations-for- the-Typological- Primacy-
Model-TPM- of- third-language- L3-transfer- Timing- of-acquisition- and-proficiency-
considered.pdf
Schwartz, B. D., & Sprouse, R. A. (2021). The Full Transfer/Full Access model and L3 cognitive states.
Linguistic Approaches to Bilingualism,11(1), 1–29. https://doi.org/10.1075/lab.20055.sch
Slabakova, R. (2017). The scalpel mo del of third language acquisition. International Journal of
Bilingualism,21(6), 651–665.
Steup, M. (1996). An Introduction to Contemporary Epistemology.Upper Saddle River, NJ: Prentice-Hall.
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 25 / 27
Gist Current models Epistemology Conclusive remarks References
References V
Steup, M., & Neta, R. (2020). Epistemology. In E. N. Zalta (Ed.), The Stanford Encyclopedia of
Philosophy (Fall 2020). Metaphysics Research Lab, Stanford University. Retrieved March 26,
2022, from https://plato.stanford.edu/archives/fall2020/entries/epistemology/
Westergaard, M. (2021). Microvariation in multilingual situations: The importance of
property-by-property acquisition. Second Language Research,37(3), 379–407.
https://doi.org/10.1177/0267658319884116
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 26 / 27
Gist Current models Epistemology Conclusive remarks References
Thanks!
Acknowledgements: Erin Leddon
Daniil M. Ozernyi
Benjamin W. Slivka Hall
2332 Campus Dr.
Northwestern University
Evanston, IL 60208
doz@u.northwestern.edu
Slides available at dozernyi.com
Daniil M. Ozernyi UNC-Chapel Hill Mar 26, 2022 27 / 27
