The sentence 'What did you wonder who bought?' is ungrammatical, while 'What did Mary say that John bought?' is fine. The best explanation for the contrast is:
AEnglish prohibits extraction of objects; only subjects can undergo wh-movement
BThe embedded question 'who bought ___' occupies the specifier of CP, leaving no intermediate landing site for 'what' to pass through on its way to the front
CWh-movement is generally prohibited across finite clause boundaries in English
DThe second sentence contains too many wh-elements for the grammar to process
This is a wh-island violation. Successive-cyclic movement requires an available specifier position at the edge of each CP phase the moving element must cross. In the grammatical sentence, no wh-element blocks the specifier of the embedded CP, so 'what' can pass through it on its way to the matrix. In the ungrammatical sentence, 'who' already occupies the embedded CP specifier, blocking 'what's' intermediate landing site — trapping it inside the island. The constraint follows from phase structure, not from an arbitrary rule about clause types.
Question 2 Multiple Choice
In the Minimalist Program, syntactic movement is best described as:
AA single direct dislocation from the base-generated position to the final surface position
BA purely phonological reordering that occurs after syntactic structure is complete
CA series of steps through intermediate phase-edge positions, checking uninterpretable features at each boundary
DA language-specific operation whose properties vary freely across languages
Successive cyclicity is the core claim: movement does not 'jump' in one step to its final landing site. Instead, a moving element must pass through the specifier of each CP (and sometimes vP) phase boundary it crosses, checking features at each stop. This architecture has empirical consequences — leaving behind 'footprints' detectable in some languages — and explains island constraints without listing them as stipulations: if an intermediate position is unavailable, the movement is blocked.
Question 3 True / False
Locality constraints on movement are language-specific rules that English has but many other languages lack.
TTrue
FFalse
Answer: False
Locality constraints follow from general principles of phase structure in the Minimalist Program, not from language-particular stipulations. Similar island effects appear across typologically diverse languages. What varies cross-linguistically is which elements trigger movement and which features drive it — but the requirement that movement proceed through accessible phase edges is universal. This is part of the appeal of the Minimalist framework: it reduces an apparent catalogue of arbitrary constraints to consequences of a more general architecture.
Question 4 True / False
A-movement (such as subject raising in 'John seems to be happy') and A'-movement (such as wh-movement) obey different locality constraints, with A-movement generally more strictly bounded to its local clause.
TTrue
FFalse
Answer: True
A-movement moves an element to an argument position (like the subject position of a higher clause) and is constrained by strict locality — it cannot cross finite clause boundaries because nominative case and phi-feature checking must happen locally. A'-movement can create long-distance dependencies spanning many clauses, as in complex wh-questions, but is blocked by island environments (wh-islands, complex NP islands, etc.) rather than strict locality. The different locality profiles reflect the different features that drive each movement type and how those features interact with phase structure.
Question 5 Short Answer
Why does the Minimalist Program require movement to proceed through successive phase edges rather than allowing a moving element to jump directly from its base position to its final landing site?
Think about your answer, then reveal below.
Model answer: Phases are domains that are 'spelled out' and become inaccessible to further syntactic operations once the phase is complete. If an element doesn't escape to the phase edge before spell-out, it is frozen inside the phase and cannot be accessed by operations in higher structure. Successive cyclicity follows from this architecture: the only way a lower element can eventually reach a high landing site is to move to the edge of each intervening phase before that phase closes. Direct long-distance displacement would require the higher operation to 'see into' an already-closed phase, which the Phase Impenetrability Condition prohibits.
This design also has the advantage of reducing computational complexity: each operation is local (within a phase), and complex long-distance dependencies arise from the composition of local steps, rather than requiring an unbounded operation that must scan arbitrarily deep into the tree.