Professional Documents
Culture Documents
Minimalist Syntax For Quantifier Raising Topicalization and Focus Movement A Search and Float Approach For Internal Merge 1st Edition Jun Abe (Auth.)
Minimalist Syntax For Quantifier Raising Topicalization and Focus Movement A Search and Float Approach For Internal Merge 1st Edition Jun Abe (Auth.)
https://textbookfull.com/product/approach-to-internal-medicine-a-
resource-book-for-clinical-practice-4-ed-edition-raj-padwal-
editor/
https://textbookfull.com/product/social-justice-multicultural-
counseling-and-practice-beyond-a-conventional-approach-heesoon-
jun/
https://textbookfull.com/product/data-merge-and-styles-for-adobe-
indesign-cc-2018-creating-custom-documents-for-mailouts-and-
presentation-packages-1st-edition-jennifer-harder-auth/
https://textbookfull.com/product/introducing-english-syntax-a-
basic-guide-for-students-of-english-1st-edition-fenn/
Transitioning to Internal Family Systems Therapy A
Companion for Therapists and Practitioners 1st Edition
Redfern
https://textbookfull.com/product/transitioning-to-internal-
family-systems-therapy-a-companion-for-therapists-and-
practitioners-1st-edition-redfern/
https://textbookfull.com/product/raising-of-microvascular-flaps-
a-systematic-approach-3rd-edition-klaus-dietrich-wolff/
https://textbookfull.com/product/european-disintegration-a-
search-for-explanations-hans-vollaard/
https://textbookfull.com/product/advanced-optimization-for-
motion-control-systems-1st-edition-jun-ma/
https://textbookfull.com/product/post-soviet-literature-and-the-
search-for-a-russian-identity-1st-edition-boris-noordenbos-auth/
Studies in Natural Language and Linguistic Theory 93
Jun Abe
Managing Editors
Marcel den Dikken, Research Institute for Linguistics, Hungarian Academy
of Sciences and Department of English Linguistics, Eötvös Loránd University,
Budapest, Hungary
Liliane Haegeman, University of Gent, Belgium
Joan Maling, Brandeis University, Waltham, USA
Maria Polinsky, University of Maryland, College Park, USA
Editorial Board
Guglielmo Cinque, University of Venice, Italy
Jane Grimshaw, Rutgers University, New Brunswick, USA
Michael Kenstowicz, Massachusetts Institute of Technology, Cambridge, USA
Hilda Koopman, University of California, Los Angeles, USA
Howard Lasnik, University of Maryland, College Park, USA
Alec Marantz, New York University, New York, USA
John J. McCarthy, University of Massachusetts, Amherst, USA
Ian Roberts, University of Cambridge, UK
Minimalist Syntax
for Quantifier Raising,
Topicalization and Focus
Movement: A Search
and Float Approach
for Internal Merge
123
Jun Abe
Tomiya, Miyagi
Japan
The work presented in this book started almost twenty years ago, when I wanted to
substantiate the idea that there are two types of minimality conditions applying to
movement. In my Ph.D. thesis, I proposed such conditions formulated in repre-
sentational terms, and when Chomsky (1995) proposed the Attract-theory of
movement, I started to think that my representational conditions could be refor-
mulated as derivational ones in terms of Attract and Move: One minimality con-
dition applies to Attract and the other applies to Move. At that time, I was aware
that this mechanism of Attract and Move would face a redundancy problem; the
consensus among the syntacticians was that the minimality condition that was
claimed to apply to Move was replaced by that applying to Attract. At the same
time, I felt that this mechanism had a conceptually attractive feature in that one
operation stands on the top of a structure and looks down to find a target phrase and
the other looks up in the structure to take this target phrase to its landing site. I also
believed that this mechanism could be given substantial evidence from covert
movement such as Quantifier Raising, which aroused a controversy with respect to
its raison d’être at that time. Although I made these ideas composed as a manu-
script, I was rather skeptical that it would attract public attention. After more than a
decade, during which the minimalist framework had changed in its technical
manifestations, it occurred to me that once Move is characterized simply as a
special case of Merge, namely internal Merge, the two operations Attract and Move
could be incorporated into the system as a prerequisite for internal Merge. This is
roughly how I came up with the mechanism of Search and Float. Though I use new
terminologies in this book, the ideas behind them have remained the same since I
started to work on this project, and the materials used to support this mechanism
were already included in my old manuscript and many of them might be felt “out of
fashion.” Nonetheless, I believe that the reader finds a new thread of ideas and
analyses in this book and hope that they inspire the reader to get interested in the
relevant topics and to conduct further investigation.
The materials reported in this book have been presented on various occasions,
most of which I have lost track of. Recently, the main idea of this book was
presented at the third annual meeting of Florida Linguistics Yearly Meeting, held at
v
vi Preface
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2 Mechanism of Search and Float for Internal Merge. . . . . . . . . . . . . . 11
2.1 Minimal Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2 Float . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3 Search and Float for Covert Movement . . . . . . . . . . . . . . . . . . . . . . . 21
3.1 Minimal Search and Float for Quantifier Raising . . . . . . . . . . . . . . 24
3.2 Covert Wh-Movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.3 Covert Oblique Movement of Wh-Phrases . . . . . . . . . . . . . . . . . . . 42
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4 Case Study I: Plurals and Reciprocals . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.1 The Cumulative Reading and Clause-Boundedness . . . . . . . . . . . . 54
4.2 The Cumulative Reading and Rigidity . . . . . . . . . . . . . . . . . . . . . . 66
4.3 Reciprocals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
4.4 Sternefeld’s (1993) Problem and Further Evidence for QR . . . . . . . 80
4.5 Clause-Boundedness and Condition A . . . . . . . . . . . . . . . . . . . . . . 93
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5 Case Study II: Multiple Wh-Questions in Japanese . . . . . . . . . . .... 99
5.1 Absorption in Japanese Multiple Wh-Questions . . . . . . . . . . . .... 99
5.2 The Scope of Multiple Wh-Phrases that Undergo Absorption . .... 107
5.3 What Is the Trigger of a Wh-Argument Being Merged
with Another . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
5.3.1 Cumulative Readings in Multiple Wh-Questions . . . . . . . . . 118
5.3.2 Pair-List Readings of Multiple Wh-Questions . . . . . . . . . . . 128
5.4 Intervention Effects and MCL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
5.5 Pair-List Readings of Plural Dono N-Phrases . . . . . . . . . . . . . . . . . 157
5.6 Apparent Pair-List Readings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
vii
viii Contents
The following examples, cited from Chomsky and Lasnik (1993) (henceforth,
C&L), illustrate how RM works:1
c. *guess [CP howi [John wondered [WHY [we fixed the car ti]]]]
(5) The possible landing sites for movement of α are possible checking positions of the
feature(s) borne by α.
1
The examples are actually taken from Chomsky (1995a), in which C&L reappears. I follow this
policy throughout this book.
1 Introduction 3
skipped this possible landing site, violating MCL. Similarly, in (3c), how could
have checked its wh-feature in the embedded Spec-CP but in fact has skipped this
position, violating MCL. What remains to be determined with this definition of
possible landing site is whether MCL excludes a case such as (3a), since it is not
clear at all what kind of feature is involved in this head movement. As pointed out
by Lasnik (1995), it is very likely that the head movement in question is just a
violation of the Last Resort Principle.
Chomsky (1993) renames MCL as the Shortest Movement Condition (SMC),
suggesting the possibility that the SMC should subsume so-called superiority cases,
as illustrated below, as well as Rizzi’s (1990) RM cases:
(8) [CP CQ [John wondered [why [we fixed the car how]]]]
2
For instance, Oka (1993a, b) reduces both RM and superiority cases to another economy con-
dition “fewest steps”, and Abe (1993) reduces both cases to MCL with the notion of possible
landing site given in (5).
4 1 Introduction
(9) a. *[why CQ [John wondered [<why> [we fixed the car how]]]]
At the stage given in (8), CP cannot attract how to derive (9b) according to the
MLC, since why is closer to CP than how and has a wh-feature that CP could attract.
Thus the only derivation in which a wh-phrase can be attracted without violating the
MLC is given in (9a). Chomsky claims that this derivation induces a semantic
anomaly with respect to the chain of why; probably, the anomaly in question is
attributed to the fact that one wh-chain marks more than one scope domain.
Although this point remained to be clarified,3 the exclusion of (9b) by the MLC and
of (9a) by a semantic anomaly correctly explains the ungrammaticality of sentence
(3c). Chomsky (1995b: 295) claims that island violations such as (9b) “involve a
longer-than-necessary move and thus fall under an approach that has sometimes
been suggested to account for superiority phenomena.” As indicated in this state-
ment, superiority violations, illustrated in (6b), will follow from the MLC
straightforwardly, since at the stage of its derivation given below:
the matrix CP cannot attract whomj due to the existence of whomi, which is closer to
CP than whomj.4
We have seen above a summary of how minimality conditions have been for-
mulated to constrain the operation Move. The main question I would like to raise
here is whether it is correct to see that MCL, given in (4), has been replaced by the
MLC, given in (7), and thus has no more role to play in the computational system of
human language (henceforth, CHL). These two minimality conditions are intended
to capture much the same type of minimality effects, hence overlapping in their
empirical coverage to a large extent, but there is a conceptually crucial respect in
which they differ in a very interesting way. Let us consider the following schematic
configuration, where [F] stands for a certain feature to check:
3
See Chomsky (2015) and Abe (2016) for recent discussions on how such a derivation as given in
(9a) is excluded under the labeling algorithm proposed by Chomsky (2013).
4
Despite the naturalness of accounting for superiority violations by the MLC, Chomsky (1995b) is
dubious about the grammatical status of the superiority phenomena. See Footnote 69 of Chomsky
(1995b) for this point.
1 Introduction 5
Metaphorically, the MLC stands in α’s point of view, saying, “I [=α] cannot
attract β to check [F] because there is another candidate for attraction, namely, γ,
that is closer to me than β.” MCL, on the other hand, stands in β’s point of view,
saying, “I [=β] cannot move to α to check [F] because there is another possible
landing site for me, namely, γ, that is closer to me than α.” I would like to argue in
what follows that this near symmetrical difference in the way the minimality
conditions are formulated is a natural consequence of applying minimal compu-
tation to two different operations working in CHL. I propose that “movement” is
factored into two sub-operations, which I refer to as Search and Float, and that these
two operations are constrained by minimal computation. Due to the nature of how
they apply, they are constrained by minimal computation in a different way, that is,
in the way that Search obeys the MLC and Float obeys MCL.5
Such a proposal will face a redundancy problem and it is usually claimed that it
is more desirable to unify similar notions or conditions. It is true that to make an
effort to avoid such a redundancy in formulating a grammar has quite often brought
some desirable consequences and has contributed a lot to the development of the
theory of grammar. There seems, nonetheless, to be no a priori reason to think that
it is always undesirable if we posit more than one kind of minimality condition such
as the MLC and MCL in a grammar. Rather, it is not implausible to claim that once
minimal computation is considered to play a central role in CHL, all operations
working in CHL should be constrained by minimal computation in its optimal form.
Chapter 2 outlines the mechanism of Search and Float, demonstrating how this
mechanism deals with typical cases of A- and A’-movement. Under the assumption
that Move is nothing but a special case of Merge, called Internal Merge, Search and
Float are characterized as prerequisite operations for Internal Merge. It is demon-
strated that the RM as well as the superiority cases shown above are explained by
the MLC applying to Search (called minimal Search), which searches down a given
structure to find the closest syntactic object carrying a relevant feature. The oper-
ation of Float is motivated by the successive-cyclicity of A- and A’-movement, as
illustrated below:
Given that Float is constrained by MCL, the reason why John and who in these
sentences must pass by the intermediate Spec-TP and Spec-CP respectively is
attributed to the fact that these positions are regarded as possible landing sites for
the Float operations applied to John and who and hence cannot be skipped.
Chapter 3 demonstrates what I believe is the best case for motivating the
mechanism of Search and Float: Quantifier Raising (QR), which is characterized as
5
Given the recent characterization of Move as simply a special case of Merge, called Internal
Merge, I characterize Search and Float as prerequisite operations for Internal Merge. See the next
section for details.
6 1 Introduction
movement for satisfying a [Scope] feature, following Abe (1993). It is well known
that some languages show so-called rigidity effects in scope interaction, as is
observed by Kuroda (1971) for Japanese:
When the surface order reflects the basic word order, as in (13a), it exhibits
so-called “rigidity” in scope interaction; that is, the structurally higher QP takes
scope over the lower QP. On the other hand, when the surface order is a derived
word order, as in (13b), either QP can take scope over the other. This indicates that
a QP that undergoes scrambling, crossing another QP, makes the scope order
ambiguous. I argue that such a rigidity effect as illustrated in (13a) is derived by
minimal Search applied at TP, which picks up only the closest phrase carrying a
[Scope] feature. I further argue that scrambling is a Search-free operation and that
this makes it possible for a lower QP to undergo Float across a higher QP. This
accounts for the scope ambiguity observed in a derived order case such as (13b).
Another well-known property of QR is clause-boundedness; consider the fol-
lowing example:
It has been standardly observed that in (14), the embedded object QP every
problem cannot take scope over the matrix subject QP someone, which thus indi-
cates that every problem takes scope in the embedded clause. I argue that this
clause-boundedness condition on QR is derived from MCL applied to a QP
undergoing Float: it prohibits such a Float operation from skipping a possible
landing site, namely the embedded TP-adjoined position, hence unable to reach the
matrix TP for Merge.
It is further argued in this chapter that in situ wh-phrases undergo covert
movement, reviving the analysis of such wh-phrases standardly assumed in the
LGB era (cf. Chomsky 1981). This analysis is motivated by demonstrating that in a
wh-in situ language such as Japanese, in situ wh-phrases show locality effects when
more than one licensing interrogative C is available and that these effects are
naturally explained on the assumption that in situ wh-phrases undergo covert
1 Introduction 7
internal Merge and that this operation is subject to the mechanism of minimal
Search and Float.
Chapters 4 and 5 are further illustrations of how QR is constrained by minimal
Search and MCL applied to Float. Chapter 4 deals with so-called plural relation
sentences such as (15a) and reciprocal sentences such as (15b):
(15a) allows what Scha (1984) calls a cumulative reading. I propose that this
reading is derived from an LF representation in which the lower plural DP is
merged with the higher one [the prisoners being merged with the women in the case
of (15a)]. I argue that this Merge operation is subject to the mechanism of minimal
Search and Float, demonstrating that a cumulative reading is not obtained when a
phrase carrying a [Scope] feature intervenes between the two plural DPs that give
rise to this reading (a violation of minimal Search) and that this reading is not
obtained from the two plural DPs that are separated by a finite clause boundary
(a violation of MCL). Further, I propose, following Abe (2000), that in a reciprocal
sentence such as (15b), each other is covertly merged with its antecedent, and that
this operation is also subject to the mechanism of minimal Search and Float. It is
demonstrated that this Merge operation does not take place across a phrase carrying
a [Scope] feature (a violation of minimal Search). Further, I argue that an apparent
violation of the clause-boundedness condition on Float, when it is applied to each
other, is attributed to the special property of the reciprocal lacking the domain to
which the semantics of reciprocity is applied; thus, in this case, a possible landing
site for such a Float operation is an argument position that could supply a reciprocal
with such a domain.
Chapter 5 deals with multiple wh-questions in Japanese, such as the following:
Based upon this fact, I argue that it is the 〈Scope〉 feature that triggers the
application of internal Merge in question. This leads us to conclude that pair-list
readings are special cases of cumulative readings; both readings are derived from
LF representations in which the two phrases involved undergo covert internal
Merge. There is a crucial difference between the two cases, however: while the
availability of cumulative readings is sensitive to the clause-boundedness condition,
that of pair-list readings does not appear to be, but rather exhibits intervention
effects. I argue that this difference is attributed to that of the quantificational nature
of the DPs involved in producing these readings: while those plural DPs involved in
producing cumulative readings may function as distributive QPs, those wh-phrases
involved in producing pair-list readings function as existential QPs. With the
assumption that existential QPs cannot take scope on their own but rather undergo
existential closure, a la Heim (1982), I argue that TP-adjoined positions are not
possible landing sites for existential QPs and hence that such QPs can cross a finite
clause boundary without violating MCL. On the other hand, I also argue that when
a wh-phrase is covertly merged with another to produce a pair-list reading, it targets
a phrase in argument position, and hence cannot skip an argument position without
inducing a violation of MCL. This derives the intervention effects observed in the
availability of pair-list readings.
Finally, Chapter 6 deals with topicalization, as illustrated in (18a), and focal-
ization such as heavy NP shift, as illustrated in (18b).
b. John met yesterday [the woman you had been talking about].
I argue that both types of movement involve satisfying the same feature, named
[S(yntactic)-Focus], which is satisfied in a left or right peripheral position, and that
whether the resulting chain functions as topic or focus is determined representa-
tionally at the LF interface. It is demonstrated that these types of movement are
subject to the mechanism of minimal Search and MCL applied to Float. The fact
that multiple applications of topicalization and heavy XP shift (where XP stands for
maximal projection of any category) in a clause are banned is derived from minimal
Search. Further, the clause-boundedness of heavy XP shift is derived from MCL
1 Introduction 9
applied to Float for satisfying [S-Focus], just like that of QR. The fact that topi-
calization is not clause-bound unlike heavy XP shift is then attributed to the
availability of Spec-CP as an escape hatch. Exploiting the notion of equidistance,
invented by Chomsky (1993), I argue that applying Float in a successive cyclic
fashion through Spec-CP can evade a violation of MCL. Further, based upon the
observation that wh-movement also undergoes Float in the same fashion and that it
may block an application of topicalization and vice versa, I argue that wh-move-
ment also involves satisfaction of [S-Focus].
It is finally argued that this analysis of topicalization and focus movement in
terms of satisfying [S-Focus] features is naturally extended to capture the locality
effects observed in such elliptic constructions as gapping, illustrated below:
Building upon Abe and Hoshi’s (1997, 1999) claim that in such a gapping
construction, the second remnant in the second conjunct undergoes rightward
movement, I argue that this property follows under the assumption (i) that the
remnant phrases in the second conjunct carry [S-Focus] features and (ii) that sat-
isfaction of this feature involves the mechanism of minimal Search and Float.
Under these assumptions, leftward movement of the second remnant (to Susan in
(19)) is prohibited by minimal Search due to the intervening DP carrying a
[S-Focus] feature (Bill in (19)). Further, the clause-boundedness observed with
gapping is attributed to the fact that the second remnant undergoes rightward
movement, which is clause-bound since MCL prohibits a Float operation for sat-
isfying a [S-Focus] feature from crossing a clause boundary, skipping a possible
landing site.
References
Abe, Jun. 1993. Binding conditions and scrambling without A/A’ distinction. Doctoral dissertation,
University of Connecticut.
Abe, 2000. How to derive the meaning of reciprocity. Minimization of each module in generative
grammar, Report for Grants-in-Aid for Scientific Research, 1–49. Graduate School of
Humanities and Informatics: Nagoya University.
Abe, Jun. 2016. Dynamic antisymmetry for labeling. Lingua 174: 1–15.
Abe, Jun, and Hoshi, Hiroto. 1997. Gapping and P-stranding. Journal of East Asian Linguistics 6:
101–136.
Abe, and Hiroto Hoshi. 1999. Directionality of movement in ellipsis resolution in English and
Japanese. In Fragments: Studies in ellipsis and gapping, ed. Shalom Lappin, and Elabbas
Benmamoun, 193–226. Oxford: Oxford University Press.
Chomsky, Noam. 1955. The logical structure of linguistic theory. Harvard manuscript, Cambridge,
MA. [Revised 1956 version published in part by University of Chicago Press, Plenum, 1975].
Chomsky, Noam. 1981. Lectures on government and binding theory. Dordrecht: Foris.
10 1 Introduction
Chomsky, Noam. 1993. A minimalist program for linguistic theory. In The view from Building 20:
Essays in linguistics in honor of Sylvain Bromberger, ed. Ken Hale and Sammuel J. Keyser,
1–52. Cambridge, MA: MIT Press.
Chomsky, Noam. 1995a. The minimalist program. Cambridge, MA: MIT Press.
Chomsky, Noam. 1995b. Categories and transformations. In The minimalist program, Noam
Chomsky, 219–394. Cambridge, MA: MIT Press.
Chomsky, Noam. 2013. Problems of projection. Lingua 130: 33–49.
Chomsky, Noam. 2015. Problems of projection: Extensions. In Structures, strategies and beyond:
Studies in honour of Adriana Belletti, ed. Elisa Di Domenico, Cornelia Hamann, and Simona
Matteini, 3–16. Amsterdam: John Benjamins.
Chomsky, Noam, and Howard Lasnik. 1993. The theory of principles and parameters. In Syntax:
An international handbook of contemporary research, ed. Joachim Jacobs, Arnim von
Stechow, and Wolfgang Sternefeld, 506–569. Berlin: Walter de Gruyter.
Heim, Irene. 1982. The semantics of definite and indefinite noun phrases. Doctoral dissertation,
University of Massachusetts, Amherst. Published in 1989 by Garland, New York.
Kitahara, Hisatsugu. 1994. Target α: A unified theory of movement and structure- building.
Doctoral dissertation, Harvard University.
Kuroda, S.-Y. 1971. Remarks on the notion of subject with reference to words like also, even or
only. In The annual bulletin of the Research Institute of Logopedics and Phoniatrics 4,
127–152. Tokyo: University of Tokyo.
Lasnik, Howard. 1995. Verbal morphology: Syntactic Structures meets the Minimalist Program. In
Evolution and revolution in linguistic theory: Essays in honor of Carlo Otero, ed. Héctor
Campos and Paula Kempchinsky, 251–275. Washington D. C.: Georgetown University Press.
Oka, Toshifusa. 1993a. Minimalism in syntactic eerivation. Doctoral dissertation, MIT.
Oka, Toshifusa. 1993b. Shallowness. In Papers on case & agreement II, MIT Working Papers in
Linguistics, ed. Collin Phillips, 255–320. Cambridge, MA: MITWPL.
Rizzi, Luizi. 1990. Relativized minimality. Cambridge, MA: MIT Press.
Scha, Remko. 1984. Distributive, collective and cumulative quantification. In Truth, interpretation
and information: Selected papers from the third Amsterdam colloquium, ed. Jeroen
Groenendijk, Theo M.V. Janssen, and Martin Stokhof, 131–158. Dordrecht: Foris.
Chapter 2
Mechanism of Search and Float
for Internal Merge
Chomsky (2004) proposes that movement is simply a special case of Merge, named
internal Merge, in which one of the syntactic objects (SOs) to be merged is picked
up from within a structure S and is merged with S. This contrasts with the original
notion of Merge, now named external Merge by Chomsky (2004), which simply
picks up two disconnected SOs. Although it is surely a desirable move to unify two
apparently distinct syntactic operations into one simple operation, i.e., Merge, that
creates an unordered set {SO1, SO2}, internal Merge is peculiar in that one target of
Merge is “buried” within a structure. In this chapter, I argue that there are two
prerequisite operations that make internal Merge possible: one is to “dig” into a
structure to find a target of Merge, which I call Search, and the other is to make this
target reach the top of the structure, which I call Float.
In order to see how this mechanism works, let us take a subject raising case for
illustration:
(1) John seems to be honest.
In the probe-goal system proposed by Chomsky (2000), the matrix T probes for
ϕ-features and finds John as its goal to check these features, as schematically shown
below:
This probing amounts to what I call Search here. The next step is to move John
to the matrix Spec-TP. A crucial question to ask here is how to ensure that John
undergoes successive-cyclic movement, passing through the embedded Spec-TP. It
has been standardly claimed that this is due to the EPP requirement on TP, but as
Chomsky (2008) explicitly notes, this is nothing but a stipulation. Abe (1997) and
Bošković (2002), on the other hand, argue that this is due to Minimize chain links
(MCL), proposed by Chomsky and Lasnik (1993), which requires that each step of
movement cannot skip possible landing sites. Given that the embedded Spec-TP is a
possible landing site for John to move to the matrix Spec-TP in (3), John must pass
through this position to satisfy MCL. Adopting this mechanism, I claim that from
the stage of derivation given in (3), John undergoes Float, which takes it to the top
of the structure in (3) for merging it with this structure and that Float is subject to
MCL, hence requiring John to be merged with the embedded infinitival TP on its
way to the top.
Search can go down a given structure as deeply as it can in principle, as wit-
nessed by the fact that the successive-cyclic A-movement involved in a subject
raising case such as (1) can be prolonged infinitely by adding raising predicates and
thereby increasing the number of embedding, as shown below:
This consideration naturally leads us to claim that the same mechanism of Search
and Float should be extended to A’-movement. Let us consider a wh-movement
case such as the following:
(6) [CP who do [TP you think [CP <who> [TP John saw <who>]]]]
(7) [CP C [TP you think [CP [TP John saw who]]]]
After this search finds who as its goal, this wh-phrase undergoes Float, and since
this operation obeys MCL, it forces the wh-phrase to pass through the embedded
Spec-CP to reach the top of this structure. In this case as well, Search can go down a
given structure as deeply as it can in principle, as witnessed by the fact that the
successive-cyclic A’-movement involved in such a wh-movement case as (5) can be
prolonged infinitely, as illustrated below:
(8) Who do you think that Mary said … that John saw?
2 Mechanism of Search and Float for Internal Merge 13
As mentioned above, Search plays the same role as probing under the probe-goal
system, but is characterized as a prerequisite operation for Internal Merge rather
than Agree, as in Chomsky (2000, 2008).2 In this sense, the present approach is
compatible with the so-called Agreeless approach, advocated by Hornstein (2009),
who argues that the operation of Agree should be dispensed with. Noting that “UG
equips FL [=Faculty of Language] with two different ways of establishing long
distance dependencies, one via Move (itself just an instance of the VCN [=Virtually
Conceptually Necessary] Merge operation) and the other via AGREE,” (p. 129)
Hornstein argues that once Move is characterized as a special case of Merge,
namely Internal Merge, it cannot be dispensed with. On the other hand, the effects
of Agree, which Chomsky (2000) proposes as an operation alternative to covert
movement, can be recaptured when a chain produced by Internal Merge has its
bottom member pronounced. I depart from Hornstein (2009), however, in claiming
that even under the Agreeless approach, Search is a necessary step to make Internal
Merge possible.
I assume that Search is minimal (hence referred to as minimal Search), just as
probing is under the probe-goal system. Let us first consider superiority cases,
illustrated in (6) of Chap. 1, reproduced below:
1
See Boeckx and Grohmann (2007) for an extensive discussion on phases and a skeptical view on
this notion as presently used.
2
See Larson (2015) for a similar view on Internal Merge.
14 2 Mechanism of Search and Float for Internal Merge
We have seen in Chap. 1 that the ungrammaticality of (9b) follows from the
Minimal Link Condition (MLC). This condition is now reduced to minimal Search,
which requires the following:
Let us consider the stage of derivation shared by (9a,b) in which the matrix C
searches for a phrase bearing a wh-feature:
Minimal Search finds whomi as its goal, but not whomj, which is located further
down the structure. Thus, (9a) is successfully derived by applying Float to whomi
after this phrase is picked up by minimal Search, whereas (9b) fails to be derived
since whomj cannot be picked up by this operation.
Let us now consider Rizzi’s (1990) Relativized Minimality (RM) cases given in
(3b,c) of Chap. 1, reproduced below:
(12) a. *Johni seems [that [TP IT is certain [ti to fix the car]]]
b. *guess [CP howi [John wondered [WHY [we fixed the car ti]]]]
We have seen that these cases are regarded as violations of MCL by Chomsky and
Lasnik (1993) and as violations of the MLC by Chomsky (1995). Here I adopt
Chomsky’s (1995) line of analysis, claiming that (12a,b) are violations of minimal
Search. Let us consider the stages of derivations of these sentences in which minimal
Search of phrases bearing ϕ-features and a wh-feature is going to apply (let us
abbreviate the former case as minimal Search[ϕ] and the latter as minimal Search[WH]:
b. guess [CQ [John wondered [why [we fixed the car how]]]]
In (13a), minimal Search[ϕ] finds it rather than John as its goal and hence (12a)
is underivable. Likewise in (13b) minimal Search[WH] finds why rather than how
as its goal and hence (12b) is underivable. The only sentences derivable from (13a,
b) are the following:
(14) a. *it seems [that [TP <it> is certain [John to fix the car]]]
b. *guess [CP why [John wondered [<why> [we fixed the car how]]]]
moves to the embedded Spec-C to satisfy its wh-feature and nonetheless moves
further to the matrix Spec-C, in which the same wh-feature could be checked.3
Note that if we assumed the phase theory, there would arise a redundancy
problem with (12a,b), since under this theory, these sentences are simply
underivable due to the fact that John and how are embedded too far down the
structure to be probed due to phase boundaries. Hence, once minimal Search is
assumed, we will face a redundancy problem that arises from the overlapping
locality effects induced by this operation and by minimal computation forced by the
phase theory. Hence, this suggests another reason for abandoning the phase theory.
2.2 Float
Given that Move is reduced to a special case of Merge, namely internal Merge,
Float may be claimed to be a residual operation of Move in the sense that it
“moves” a phrase down in a structure to its top. Thus, Float must be applied
upward, as is implied by the name of this operation, just like movement in its
traditional sense. Float differs from Move, however, in that it simply takes an SO
upward to a given dominating category without “attaching” it to the latter category;
it is the Merge operation that takes care of this attachment. To put it in other words,
the traditional notion of Move is now factored into two separate operations: Float
and Merge. Thus, Float of α may be defined as an operation comparable to Move
without merging α with a target phrase. I have proposed that this operation, together
with minimal Search, is a prerequisite for internal Merge. Under this conception,
the following condition is operative in the relationship between Float and Merge:
way of Float.
Thus, when the present mechanism of Search and Float is compared with
Chomsky’s (2008) system of probe-goal and internal Merge, it will be clear that its
peculiarity resides in this requirement.
It has usually been assumed in the minimalist framework that the upward
property of movement is derived from the Extension Condition imposed upon
(internal) Merge (cf. Chomsky (1993)), which requires that structure be extended or
the No Tampering Condition (cf. Chomsky (2008)), which also has the effect of
(internal) Merge taking place at the root. But these assumptions are not compatible
with the present system of Search and Float. Recall that I have argued above that
3
See Abe (2016) for an attempt to derive the ungrammaticality of (14a,b) from failure of labeling
under the labeling algorithm proposed by Chomsky (2013). Note that (14a,b) are independently
ruled out due to the fact that John fails to check its Case-feature in (14a) and how cannot stay
in situ in (14b), but these are simply accidental properties of these sentences.
16 2 Mechanism of Search and Float for Internal Merge
b. [CP who do [TP you think [CP <who> [TP John saw <who>]]]]
Under the present assumptions, John and who in (16b) and (17b), after they are
picked up by minimal Search, each undergo Float to the top of the structure, and
they must stop by the embedded TP and CP for merging with them in order to
observe MCL. These Merge operations, however, clearly violate the Extension
Condition or the No Tampering Condition. One might claim that these conditions
are imposed not upon (internal) Merge itself but rather upon minimal Search.
Although it is true that this assumption is compatible with the present mechanism of
Search and Float, this alone will not guarantee the upward property of movement.
Let us consider the following example:
(18) Who do you think that the man who said that Mary was smart saw?
This sentence is a variant of (17a) where John is replaced by the man who said
that Mary was smart. In this case, we want to claim that who undergoes
successive-cyclic A’-movement in exactly the same way as that in (17a) does; that
is, it stops by the Spec of the CP complement of think for internal Merge without
violating MCL. Thus, it must be the case that the Spec of the CP complement of
said is not regarded as a possible landing site for the Float operation in question.
This will follow from the assumption that Float must be upward. There will be a
number of ways to deduce this upward property of Float, but I will simply assume
the following without further argument:
Given this condition, Float of who in (18) can go though the CP complement of
think, since the latter dominates who, which thus enables who to merge with this
complement clause on its way to the matrix CP. On the other hand, this Float
operation does not go through the CP complement of said, since the latter does not
dominate who, which thus makes it impossible for who to merge with this
2.2 Float 17
complement clause. This derives the fact that the Spec of the CP complement of
said does not count as a possible landing site for who in (18).
In addition to the assumption that Float is subject to MCL, it will not be
unnatural to claim that Float is also subject to island conditions, whose typical
examples are given below:
b. *Whoi did John come back before I had a chance to talk to ti?
(ibid.: 497)
(20a) illustrates a case of relative clause island violation and (20b) a case of
adjunct clause island violation. In these cases, it can be claimed that although
minimal Search finds the wh-phrase who as its goal, Float cannot take this wh-
phrase to the top of the structure due to the islandhood of relative and adjunct
clauses. In this book, I will not address the question of how the islandhood of a
given configuration is determined and why this restriction is imposed upon Float.
There is good reason to regard island conditions as applying to Float rather than
Search.4 This is concerned with the fact, observed by Saito (1985), among others,
that scrambling, a movement operation whose main role is to change word order, is
also subject to these conditions, as illustrated below with Japanese examples:
rakutansiteiru (koto)
is-depressed fact
4
However, see Goto (2015) for the contrary claim that it is Search that is subject to island
conditions and that islandhood is derived from lack of a label.
18 2 Mechanism of Search and Float for Internal Merge
hence allowed to pick up any phrase to apply Float to. Then, the fact that scram-
bling shows island sensitivity follows from the assumption that it is Float rather
than Search that is subject to island conditions.
It is predicted under the present characterization of scrambling as a Search-free
operation that it will not show wh-island effects such as shown in (12b), since these
effects are derived as a result of minimal Search violations. This prediction is in fact
borne out, as Saito (1985) notes that scrambling shows very weak effects of wh-
islands, as illustrated below:
(22) ?[Sono hon-o]i [John-ga [Bill-ga dare-ni ti ageta ka] siritagatteiru (koto)
In order to derive this sentence, we can just pick up sono-hon-o ‘that book-ACC’
and apply Float to this phrase. Since no islands in the sense relevant here intervene
in this operation, Float can successfully take sono hon-o to the top to the structure
for applying Merge.5
To sum up this chapter, I have argued that internal Merge requires two additional
operations, unlike external Merge: Search and Float. Search finds a target of Merge
within a given structure, and this operation is minimal in that it must pick up the
closest phrase that carries a relevant feature. Float takes this targeted phrase to the
top of a given structure for Merge, and this operation is subject to island conditions
and MCL, which requires that it not skip possible landing sites.
References
Abe, Jun. 1997. What triggers successive-cyclic movement. In ‘Is the logic clear?’: Papers in
honor of Howard Lasnik, University of Connecticut Working Papers in Linguistics 8, ed.
Jeong-Seok Kim, Satoshi Oku and Sandra Stjepanović, 1–20. Cambridge, MA: MITWPL.
Abe, Jun. 2015. The nature of scrambling and its resulting chains: Operator or mediator of various
constructions. Ms.
Abe, Jun. 2016. Dynamic antisymmetry for labeling. Lingua 174: 1–15.
Boeckx, Cedric, and Kleanthes Grohmann. 2007. Remark: Putting phrases in perspective. Syntax
10: 204–222.
Bošković, Željko. 2002. A-movement and the EPP. Syntax 5: 167–218.
Chomsky, Noam. 1993. A minimalist program for linguistic theory. In The view from building 20:
Essays in linguistics in honor of Sylvain Bromberger, ed. Ken Hale, and Sammuel J. Keyser,
1–52. Cambridge, MA: MIT Press.
5
It is also predicted that scrambling is free from superiority effects, but this prediction is not fully
borne out: while clause-internal scrambling does not exhibit such effects, long-distance scrambling
does show such effects, as observed by Takahashi (1993). Here I will not dwell on why this
contradictory property holds, but see Abe (2015) for relevant discussion.
References 19
Chomsky, Noam. 1995. Categories and transformations. In The Minimalist Program, Noam
Chomsky, 219–394. Cambridge, MA: MIT Press.
Chomsky, Noam. 2000. Minimalist inquiries: The framework. In Step by step: Essays in honor of
Howard Lasnik, ed. Roger Martin, David Michaels, and Juan Uriagereka, 89–155. Cambridge,
MA: MIT Press.
Chomsky, Noam. 2004. Beyond explanatory adequacy. In Structure and beyond: The cartography
of syntactic structures, vol. 3, ed. Adriana Belletti, 104–131. Oxford: Oxford University Press.
Chomsky, Noam. 2008. On phases. In Foundational issues in linguistic theory: Essays in honor of
Jean-Roger Vergnaud, ed. Robert Freiden, Carlos P. Otero, and Maria Luisa Zubizarreta,
133–166. Cambridge, MA: MIT Press.
Chomsky, Noam. 2013. Problems of projection. Lingua 130: 33–49.
Chomsky, Noam, and Howard Lasnik. 1993. The theory of principles and parameters. In Syntax:
An international handbook of contemporary research, ed. Joachim Jacobs, Arnim von
Stechow, and Wolfgang Sternefeld, 506–569. Berlin: Walter de Gruyter.
Fukui, Naoki. 1993. Parameters and optionality. Linguistic Inquiry 24: 399–420.
Goto, Nobu. 2015. Restricting n to two: When Merge requires search. Ms., Toyo University.
Hornstein, Norbert. 2009. A Theory of syntax: Minimal operations and universal grammar. New
York: Cambridge University Press.
Huang, C.-T. James. 1982. Logical relations in Chinese and the theory of grammar. Doctoral
dissertation, MIT.
Larson, Bradley. 2015. Minimal search as a restriction on Merge. Lingua 156: 57–69.
Rizzi, Luigi. 1990. Relativized Minimality. Cambridge, MA: MIT Press.
Rizzi, Luigi. 2006. On the form of chains: Criterial positions and ECP effects. In Wh-movement:
Moving on, ed. Lisa Lai-Shen Cheng and Norbert Corver, 97–133. Cambridge MA: MIT Press.
Saito, Mamoru. 1985. Some asymmetries in Japanese and their theoretical implications. Doctoral
dissertation, MIT.
Takahashi, Daiko. 1993. Movement of wh-phrases in Japanese. Natural Language and Linguistic
Theory 11: 655–678.
Chapter 3
Search and Float for Covert Movement
In this chapter, I argue, following May’s (1977) original idea about Quantifier
Raising (QR), that since quantifier phrases (QPs) and in-situ wh-phrases have the
property of taking scope, they cannot obtain proper interpretations at the LF
interface unless they appear in peripheral positions where they can take scope. This
forces QPs and in-situ wh-phrases to undergo covert movement. I argue that such
covert movement is best formulated as feature-triggered movement and thus is
amenable to the mechanism of minimal Search and Float.
Let us follow Abe (1993) in assuming that quantifiers such as everyone and
someone have the feature [Scope], and that this feature is satisfied by moving its
carrier to a scope-taking position, typically a TP-adjoined position. Thus, in sen-
tence (1a) below, every boy carries a [Scope] feature and undergoes covert
movement to satisfy this feature, as shown in (1b):
(1) a. John hates every boy.
[Scope]
This is one illustration of the way QR is incorporated into a system where the
trigger of movement is attributed to features. If we assume that TP-adjoined
June 29.—This morning we sent for a carriage and sat out to see the city
and make some visits. Buffalo, although suffering with all our cities in the
stagnation of trade, seems to be doing a great deal of business. The rows of
shops, and handsome ware-houses, seem to contain every article necessary
for comfort or luxury. It is a larger city than Rochester, but has not its air of
elegance and neatness. The town was burnt by their neighbors, the
Canadians, in 1814, but has since been rebuilt. The streets are wide and
airy, Maine street, the principal avenue, is more than a mile in length. The
churches are neat buildings, one of them, a catholic, promises, when
finished, to be handsome. The court house is a solid well built edifice
having pillars up to the roof. The markets are very good also. The city is
well situated upon ground rising gently from the lake, the upper part being
covered with handsome private dwellings, which thus obtain fine views of
the lake and surrounding country, and secure for themselves room for their
gardens which are very prettily laid out. There is here also a military station
for the United States troops, whose barracks, comfortable brick buildings,
are built around the parade ground and surrounded by a good wall. Our
friend’s cottage was upon elevated ground looking down upon the green
Niagara river, and enjoying a view of the lake in front, and behind an extent
of country covered with the untamed forest. It was the first time I had seen
a forest landscape, and I looked with much interest upon this vast plain of
green leaves reaching to the distant horizon; a smoke curled in one spot
telling of some settler clearing his way through the green wood. The
handsomest private dwellings here do not affect the Gothic or Grecian,
which had prevailed along our road, but were substantial square stone or
brick buildings, having a marble portico in front, an cupola upon the top,
surrounded by a fancy railing. Our drive around Buffalo was very
interesting, and we wondered, as we marked such a mass of solid buildings,
and depots of articles from every region in the world, and such throngs of
human beings deposited in a wilderness, but a few years redeemed from the
Indian, the buffalo, and the bear. What industry, what energy, has been
employed to bring hither all these materials. Buffalo is a frontier town, and
grand portal of the west, through which is flowing a constant stream of
travellers and emigrants. This mixture of all nations in the streets, give them
an unique appearance. Here you see the Indian beau with his tunic bound
with a crimson sash, his hat surrounded by a circle of feathers; his deer skin
pantaloons richly embroidered in barbaric patterns, while ribbons and
tassels swing out from his dress at every step. After him will pass a band of
United States soldiers; then a rough back-woodsman, upon a horse looking
as wild as himself, its uncut mane and tail waving in the wind as he gallops
violently through the streets. Then follows a party of comical German
emigrants; a scarlet clad British officer; a Canadian; a Frenchman; a wild
looking son of Erin; a sturdy ruddy, gaiter legged English farmer; a
Tonawanda squaw with her papoose upon her back; and lastly the dainty
lady traveller with her foreign abigal, and fantastically dressed children.
Among the crowd I observed a curious figure—a one legged negro, wearing
an old uniform coat with ruffled cuffs, ringing a bell most energetically. The
old English custom of sending a bell-man to proclaim the loss of any
article, prevails here, as in some of our other towns, I believe. ‘What is lost,
Sambo?’ inquired a person. ‘Your wits, massa,’ he replied quickly, setting
his juvenile train off in a fit of laughter. To another inquirer, he replied, ‘My
leg is lost, don’t you see’ holding up the stump. He is, I suppose a
privileged wit, who, if he cannot set the table, no doubt does the street in a
roar. The Buffalonians are a gay social people. The unamiable fashion of
exclusiveness being very little known here, for, living where the population
is continually changing and where strangers are constantly claiming their
hospitality, they have acquired an easy unsouciant manner, and are ever
forming social meetings to entertain the stranger. Our letters procured for us
much kind attention, and we had an opportunity of witnessing this free
hospitable spirit. In the afternoon one of our friends called, and we drove
down where a fanciful yacht awaited us, and a pleasant party of ladies and
gentlemen, for the purpose of taking us over to the ruined fort opposite the
city. This is a favorite picnic haunt of the young citizens. Fort Erie is upon
the Canadian shore, opposite Buffalo, just at the point where the Niagara
river runs out of lake Erie. It was destroyed during the war of 1812.
I have scarcely enjoyed any thing so much as that sail over Lake Erie.
The lake is here five or six miles broad. The water rushes swiftly past, as if
eager to accomplish its glorious destiny of plunging over the rocks of
Niagara, there to be a spectacle which nations come from afar to gaze upon.
We caught the excitement which seemed to animate the water, as we were
tossed upon its wavelets with quick, gay, tilting motion; and gazed with
much delight at the novel objects around us. The city, with its numerous
domes and spires; the bright Niagara rushing and gurgling at a rapid rate
over the ledge of rocks which once was Erie’s barrier ere the waters burst
their bounds—the gulls wheeling above us, or floating upon the waves; and
above all, that immense lake, that mighty mass of sparkling emerald water,
stretching far into the mysterious west. The air, breathing from the fresh
forest and cool lake, was so refreshing that I was almost sorry when we
reached the shore. Landing upon a sandy beach, we repaired to the fort,
where under the shadow of a ruined wall, we seated ourselves upon the
green sward, and while refreshing ourselves with the contents of our
provision baskets, our discourse fell upon the hapless fate of those whose
blood had dyed the fair turf around us; or upon other scenes which occurred
during that border war. But now all this is over; conqueror and vanquished
are both beneath the ‘clod of the valley’; the echo of the war trump has died
away; the green earth smiles again as peacefully as if it had never drank the
blood of the dying, and wall, and bastion, are fast crumbling into their
parent elements. The lake, the sky, the shore, are no longer vexed with
sights and sounds of strife. Alas! whence come wars and fighting among
us? Must these things always be? Must earth’s children ever thus hack and
tear each other? And we who are brethren, whose homes are in sight upon
either shore of this bright lake, can we not dwell in unity? They who have
opposite creeds, who differ in dress, in manner, in language, may and will
rival, dislike, detest, fight and exterminate each other; but we, who are sons
of the same father, who speak the same tongue, Oh, must we be ever thus at
enmity?
[7] Though ages long have passed
Since our fathers left their home;
Their pilot in the blast
O’er untravelled seas to roam,—
Yet lives the blood of England in our veins!
I have said this is a favorite place of resort, and here a party of gay
young people came to avoid the noise of the city, and spend a quiet day
with their books and work, upon the fourth of July. Their little feast was
spread under the shade of the fortress, and they were in the act of drinking
to the day, when they were suddenly taken captive by a band of English
soldiers. It was at that unhappy time when Canada was disturbed by
revolutionary projects, and it was naturally imagined they had come there
purposely to insult them. It was an imprudent frolic, and they paid dearly
for it; they were marched off three miles to a military station, where, after
being fully examined and no signs of revolution being found upon them,
they were suffered to depart and return as they best might. I relate the
anecdote to show how easily we may mistake each other’s motives, and
how soon ill-blood may be brewed between those who are suspicious of
each other, and ready to take offence.
While we were thus discoursing, the sky grew gradually dark, and a veil
of blackness was let down over the lake, giving token of a thunder shower.
We were soon in the boat which tossed very much, but we had able young
seamen who landed us safely just as the sun, bursting forth, smiled at our
idle fears. An evening of social pleasure ended our agreeable day.
June 30th.—This morning we were again employed in rambling about
the city. The situation of Buffalo is calculated to make it a great commercial
mart. It is upon the high road to the west, and will command much of the
business of the lakes, while the great Erie canal connects it with the
Atlantic. This canal is indeed a ‘herculean achievement.’ It is three hundred
and sixty-three miles in length, forty feet wide, and four deep; contains six
hundred and eighty-eight feet of locks; is crossed by several fine aqueducts;
and all this was completed in eight years. There are other canals connected
with it. This great artery, bringing up the produce of Europe to the west,
through this city, must increase its prosperity[8] and population.
At twelve o’clock this morning embarked in the steamboat Constellation
for Chicago, through lakes Erie, St. Clair, Huron and Michigan, a distance
of twelve hundred miles, for which we are to pay twenty dollars, ten each.
The wharves as we left them presented a busy scene. We counted forty
steamboats and canal boats, beside several large vessels. Among the latter
was the Queen Charlotte, a stately ship of war belonging to Canada, but
degraded to the ignoble fate of a Buffalo trader. She had, it is true, lost some
of her original brightness ere thus fallen, for she had been twenty-three
years under water, having been sunk in a naval fight on Lake Erie, and
lately raised. The wharves were loaded with produce and merchandize,
while carts, boats, and men, were loading and being unloaded.
We left Buffalo with regret. Its majestic river and noble lake—its back
ground of forests, gay streets, and social people, have left a vivid and
pleasing picture upon our memories. A fine pier, or breakwater as they call
it, of solid mason work extends 1100 feet, protecting the wharves from the
waves. A light house stands upon the end of the pier. When the city had
completely faded into the distant horizon we turned our gaze on our
companions. Upon one corner of the deck was a promiscuous heap of
chairs, children, pots, kettles, men and women, being a family moving west.
That old man with a cocked hat, and large metal buttons, the young man in
a blue frock, and women with embroidered stomachers and indescribable
caps, sitting upon a pile of strange looking articles of husbandry, and huge
unwieldy chests, is a band of emigrants from central Europe. A party of
English gentleman from Canada were there, bound upon a hunting
expedition to Wisconsin—another of Buffalo young men, were going to
while away the summer months in a fishing excursion upon Lake Superior,
a long light skiff being part of their travelling luggage. There were also
tourists for pleasure, information and health like ourselves, and some few
going to inspect lands which they had bought unseen. Our steamboat is a
very fine one although not of the first class. There is a handsome saloon for
the ladies surrounded by a circle of state-rooms opening upon the deck—
below are the eating rooms and gentlemen’s cabin, the whole fitted up with
comfort and elegance. There are about 53 steamboats upon lake Erie, some
of them of six hundred tons, and fitted up with every luxury and elegance,
many costing from $15,000 to $120,000 each. They are built upon a fine
model, and are well finished. The upholsterer’s bill sometimes amounts to
$4,000. They are generally built very strong to resist the waves that run
high here. The complement of men for one of these boats amounts to 40;
the captain receiving $100 a month. After an excellent dinner we ascended
to the promenade deck which, like our Hudson river boats is the uppermost
deck, surrounded with seats. We were out upon lake Erie, and gazed around
us with wonder and delight. The water was a fine dark green, which as the
wind was high, was tossed in waves crested with white foam, or sparkling
spray. The shores were in some places low and wooded, alternating with
gentle elevations, at whose foot ran a line of yellow sand—a sky of purest
azure dotted with fleecy clouds was above. What a lovely scene—
asks Sir Walter. This lake however is rather larger than his Scottish lake, it
being 290 miles long. It has the character of being the most tempestuous of
all the lakes, a fact we were soon able to verify, for in the afternoon the
wind increased to a gale, and the waves dashing against our vessel gave us
each time a shock as if she had struck a rock.
The ladies soon began to feel the effects of such tossing, and one after
another retired to their berths quite ill. Forty-five miles from Buffalo we
stopped at the town of Dunkirk, which is the termination of the New York
and Erie rail road. It commences at Hudson river 25 miles above the city of
New York, a distance of 450 miles from its end. This town, under these
circumstances, is rising rapidly. It has a fine circular bay having two
projecting points which protect it, one and a half miles across—and is one
of the best harbors upon the lake. There is also a pier within the shelter of
which five large schooners were moored. We observed a rail road depot
ready for the future engines and train—a church, tavern and a few stores.
Several little boys came on board with pails of cherries for sale, which they
disposed of at four cents a quart. Here we landed a passenger, an inhabitant
of Dunkirk, who, during the voyage, had been vaunting the advantages of
his town. The day would soon come, he said, when he should no longer
resort to Buffalo for his goods, as the new rail road would bring all the trade
to Dunkirk. Darkness drove us to our state room, which we found replete
with every convenience—a circumstance much to our satisfaction as we
were to spend a week in it ere we reached our destined haven. I would
recommend you if you ever travel this way, to choose, as we did, a state
cabin looking towards the shore, for these boats stop at every considerable
town, and of course keep near the coast. In consequence of this
arrangement, we could, if inclined, sit in our cabin, and through the open
door, or window, behold the scenery at our ease; while those upon the
opposite side, gazed out upon an uniform waste of waters without a shore. I
thus obtained a sight of the town of Erie where we stopped during the night.
Aroused by the noise, I looked from my window and saw the town
distinctly by clear starlight. This town is in Pennsylvania, and is the
termination of the Pittsburg and Erie canal. In the canal basin, beside canal
boats, I saw a large steamboat and several schooners. Presque Isle defends
the harbor. There was a large hotel brightly illuminated, and some stage
coaches, awaiting the arrival of passengers. Erie stands upon a high mass of
Schistose rock surmounted by a stratum of clay—the whole forty feet above
the lake.[9] There is said to be here a neat court house, and several pretty
houses surrounded by trees—the streets are at right angles, and the trade
considerable. There was a bridge spanning the canal, which I hoped was the
one where the revered La Fayette was feted. It was formed into a large tent
by sails and flags, which had waved in the battle upon the lake, under which
was a fine collation. Several ships of war have been built here. You will
surely give me credit of being a first rate correspondent when I leave my
slumbers to collect items for your amusement and edification.
This man’s history interested us much, and I will relate it for your
edification. He was a native of our city of New York, one of a large family
straightened for means. While quite young he had married, and struggled
for years to support his family respectably, but sickness and ‘bad times’
rendered his lot a gloomy one. Hearing so often of the happiness and
prosperity of ‘the west,’ he resolved to remove thither, and accordingly
bought a tract of land upon St. Clair river, then farther west than it now is.
He came here twenty years since, with a wife and several young children,
and a mere trifle in money. A little village has now risen around him, of
which he is the owner. He has built a good tavern for travellers, which he
rents out; has erected a saw-mill; a few shops and houses, and a little
church. His children are married and settled around him; and he is, as he
expressed himself, ‘independent of the world.’ Once a year he goes to New
York or Buffalo, to purchase goods for his shop. How much better is this
state of things than to remain, struggling for a morsel, among the hungry
crowd of a large city. I asked him if he never repented renouncing a city
life. ‘No, indeed!’ he answered—‘I go there once or twice a year to transact
business, but hurry away, for I feel as if in prison. I want elbow room, and
never breathe free until threading my green lakes and vast forests again. I
am glad to leave such fictitious existence, where each man models his
conduct upon that of his neighbor, and dare not act as his spirit prompts
him.’ We had passed into St. Clair river, and about sun-down dropped this
man and his goods at his little village, which was seated upon a green slope,
cut out of the forest, upon the Michigan shore. The houses were surrounded
by little gardens and seemed comfortable.
The sign of the village inn was swinging in the summer breeze; a
traveller had just alighted from his horse in front of the piazza, and the
steam from his mill was rising high above the trees tinted purple in the
evening light. From a shop door a young man, probably his son,
accompanied by a neighbor, stepped forth to greet him; while, from the
honeysuckle-covered porch of a neat cottage a woman, whom I fancied his
wife, was looking eagerly out to watch his approach. Every thing denoted
industry, cheerfulness, and independence.
Soon after leaving the village of Clay, we observed a ship at anchor near
the shore, quite a picturesque object. It proved to be the Milwaukie, a ship
of three hundred tons burthen, bound from Buffalo to Chicago. It was
waiting for wind, or steam, to enable it to enter lake Huron, as this lake
pours into the river St. Clair with so strong a current, that vessels can
seldom stem it without a strong wind. She was soon attached to our
steamboat, and we both passed swiftly along. What a superb western sky!
The sun has long left us, and yet we scarcely miss its light, so golden and so
brilliant is the mantle he has left behind him. It is nearly nine o’clock, and
yet I can see to write this, but fatigue drives me to my cabin, and forces me
to say adieu until to-morrow.
July 3d.—Still in the river St. Clair. We stopped some hours in the night
at Newport, to take in a supply of wood. The captain purchased eighty cords
at $1,50 a cord. He told us it was his opinion the steamboats upon these
waters would soon be obliged to burn coal, although surrounded by such a
world of trees, as there is so much time wasted in stopping for it. I did not
regret our detention, as I was anxious to lose no part of a scenery to me so
novel and pleasing. This is a beautiful river about sixty miles long, and half
a mile broad, having several little towns upon it. Cotrelville and Palmer we
had also passed in the night; the latter a thriving place, from which a rail
road is contemplated to Romeo, twenty-six miles, there to meet the Shelby
and Detroit rail road. A communication will thus be continued with Detroit
through the winter. The country upon the Canadian shore is wild and
uninhabited, while the Michigan side of the river is frequently adorned with
fields of grass or wheat, or thrifty orchards. The houses are plain, but
seemed surrounded by every comfort. Our course ran quite near this shore,
so close, that I might fancy myself transported into the midst of a farm yard,
with all its morning business going on. A pretty white wood house is before
me now, surrounded by fields and barns, having a row of cherry trees in
front whose fruit is glistening red in the morning sun. In the barn yard a
man is chopping wood, to cook the breakfast, I suppose—another is busy
hoeing in a potatoe field—a boy is leading a horse down to the river for
water, while numerous other children are arrested in their play and stand
open mouthed gazing at us—ducks are dabbling in the wavelets—pigs are
rooting up the turf—a flock of geese are running down the bank at us with
beaks and wings extended in a warlike attitude—while a sober cow chews
her cud under a large hickory nut tree. The next moment all is gone, to give
place to the silent groves of oak, maple and ash. Upon a long narrow island
near the Canadian shore, my eyes were attracted by what seemed a row of