(image source: NYU)
1: Linearization as a Core Problem of Syntactic Theory
One of the most persistent assumptions in late twentieth-century generative grammar has been that syntax is fundamentally hierarchical and that linear order is imposed only at the point of externalization, commonly associated with Phonetic Form (PF). On this view, syntax generates abstract structures consisting of asymmetric relations such as c-command and dominance, while word order is treated as a secondary, interface-driven phenomenon. Professor Richard Kayne’s work challenges this architectural division at its foundations. The central claim advanced here is that linear order is not an accidental by-product of phonological realization but an intrinsic property of syntactic computation itself.
The problem of linearization is, therefore, not a peripheral technical issue. It cuts to the heart of what syntax is taken to be and how grammatical knowledge is represented in the human mind. If syntax itself determines linear order, then syntactic theory must explain why languages display the orders they do and, equally importantly, why certain logically possible orders are unattested. Kayne’s anti-symmetry program is a sustained attempt to answer precisely these questions.
2: From Early Generative Syntax to the Externalization Hypothesis
In the earliest phases of generative grammar, linear order was treated as a basic component of syntactic representation. Phrase structure rules explicitly encoded order, and transformational operations were defined over ordered strings. Beginning in the 1980s, however, a different conception emerged. Influential work suggested that hierarchical relations were primary and that order could be factored out, introduced only when structures were mapped to sound.
This shift was motivated in part by the desire to simplify the core syntax and to locate variation at the interfaces. If linear order were introduced late, then syntax could be invariantly hierarchical across languages, with differences in word order attributed to language-specific rules of externalization. Kayne argues that this move, while conceptually attractive, weakens explanatory power. By removing order from syntax, it becomes difficult to account for the strong and systematic asymmetries observed across languages.
3: The Anti-Symmetry Hypothesis and the Linear Correspondence Axiom
Kayne’s 1994 monograph introduced the Anti-Symmetry Hypothesis, formalized through the Linear Correspondence Axiom (LCA). The core idea is that hierarchical asymmetric relations map directly onto linear precedence relations. In effect, if a syntactic object asymmetrically c-commands another, it must precede it in linear order.
A crucial consequence of this mapping is the impossibility of mirror-image syntactic systems. A grammar that generated both a structure and its mirror image would violate anti-symmetry. The result is a universal linear order of constituents: specifier precedes head, and head precedes complement. Importantly, this is not a parameter but a structural consequence of the syntax itself.
Anti-symmetry therefore excludes rightward adjunction, multiple specifiers per head, and freely ordered complements. What might initially appear as stipulative restrictions instead emerge as principled consequences of a tightly constrained mapping between structure and order.
4: Anti-Symmetry and the Rise of Cartography
The cartographic approach to syntax, which posits richly articulated hierarchies of functional projections, finds a natural ally in anti-symmetry. If each head licenses at most one specifier, then the proliferation of distinct projections becomes a structural necessity rather than an analytic indulgence. Differences in interpretation correspond to differences in structural position, not to optional rearrangements within a flat structure.
Although cartography goes beyond anti-symmetry in claiming universal skeletal structures, the two approaches converge on the idea that syntactic variation is tightly constrained. Both reject the view that languages differ freely in basic structural organization.
5: From Axiom to Theorem
In its original formulation, the LCA functioned as an axiom. In more recent work, Kayne has argued that this is unsatisfactory. If syntax is part of a biological cognitive system, then its fundamental properties demand explanation. The question is no longer whether anti-symmetry holds, but why it should hold.
This shift aligns anti-symmetry with the broader goals of the Minimalist Program: to derive grammatical properties from more general principles of computation and cognition. Rather than stipulating Spec–Head–Complement order, the task is to show how such an order follows from independently motivated mechanisms.
6: Why Spec–Head–Complement?
Anti-symmetry alone allows two linearizations compatible with hierarchical asymmetry: Spec–Head–Complement and Complement–Head–Specifier. Kayne’s proposal is that additional constraints, related to probing and feature valuation, select the former.
The technical details of this proposal are complex, but the guiding intuition is clear. Heads and specifiers must occupy distinct positions relative to complements, and the grammar consistently chooses one configuration. This choice is not arbitrary but reflects deeper properties of syntactic computation.
7: OV Languages and the Japanese Case
At first glance, languages such as Japanese appear to falsify Spec–Head–Complement order. If objects precede verbs, does this not imply head-final structure? Kayne argues that this inference is mistaken. If the object were merged as a complement to the verb, anti-symmetry would indeed be violated. The conclusion, then, is that the object is not in complement position.
Instead, objects in OV languages occupy specifier positions reached via movement. This analysis extends the widely accepted hypothesis that subjects originate within the verb phrase and move to higher positions. Object movement, on this view, is no less fundamental than subject movement.
8: Cross-Linguistic Evidence for Object Movement
Evidence from German, Dutch, and West African languages reinforces this conclusion. Constructions such as the IPP pattern in Germanic languages yield word orders incompatible with simple complement–head analyses. Similarly, object placement relative to negation in various languages indicates that objects must occupy positions higher than the verb.
These patterns converge on a single conclusion: surface OV order is derived, not basic. What varies across languages is not the underlying structural order but the extent and nature of movement.
9: Movement as the Default for Arguments
The idea that arguments reach their surface positions via movement has a long history in generative grammar. Kayne extends this logic uniformly to subjects, objects, and even objects of prepositions. Prepositions, on this view, probe for their arguments rather than merging with them directly.
This reconceptualization of argument structure further erodes the traditional distinction between complements and specifiers as primitive notions. What matters is not where an element is merged, but how it is integrated into the derivation.
10: Copies, Pronunciation, and Linear Order
Kayne expresses cautious sympathy for copy theory, according to which movement creates multiple copies of an element, only one of which is pronounced. While this theory captures important generalizations, it leaves unanswered questions. Why are lower copies typically silent? Why do some constructions allow multiple pronunciations while others do not?
These questions suggest that linearization interacts with movement in ways not yet fully understood. Anti-symmetry sharpens the problem by insisting that linear order is determined internally, not imposed at PF.
11: Free Word Order and Morphological Richness
Languages often described as having free word order exhibit extensive movement constrained by case morphology and information structure. Comparative work shows that freedom of order is always limited and systematic. Morphological richness facilitates movement but does not abolish structural constraints.
This observation supports the anti-symmetry view that all languages share a common underlying order, with variation arising through derivational complexity.
12: The Elimination of the Head Parameter
If Spec–Head–Complement order is universal, the classical head parameter must be abandoned. The task of acquisition then shifts from setting a parameter to learning the movements that derive surface order. This reconceptualization aligns with empirical findings that children acquire word order through exposure to patterns of movement rather than by selecting between abstract options.
13: Flat Structures and Branching
Proposals for radically flat structures raise challenges for anti-symmetry. Kayne argues that true flatness is unlikely to be a point of parametric variation. Apparent flatness can often be reanalyzed as the result of unconventional but still hierarchical mergers. Binary branching, on this view, remains a universal property of syntax.
14: Anti-Lexicalism and Syntax Within Words
Anti-symmetry extends naturally to word-internal structure. If syntax determines linear order, there is no principled reason to restrict it to phrasal domains. Kayne rejects strict lexicalism, arguing instead that morphemes are the true terminals of syntax. Greenberg’s universals linking affix order and syntactic order support this continuity between words and phrases.
15: Open Questions and the Future of Anti-Symmetry
The anti-symmetry program remains an active area of research. Questions persist about whether anti-symmetry applies fully within syntax or only at externalization, whether weakened versions are empirically necessary, and how determiner structure should be mapped cartographically.
These debates do not weaken the program; they testify to its vitality. By forcing explicit engagement with linear order, anti-symmetry has reshaped the agenda of syntactic theory.
16: Conclusion
Richard Kayne’s work on linearization challenges linguists to rethink the architecture of grammar. By insisting that linear order is a core syntactic property, anti-symmetry restores explanatory depth to word order phenomena and constrains the space of possible grammars. Whether ultimately confirmed or revised, the anti-symmetry hypothesis has irrevocably altered how syntax is theorized, placing linear order at the center of grammatical explanation.
