English tongue-twisters constitute phonologically constrained utterances whose production becomes difficult under conditions of repetition and increased speech rate. The present study investigates their phonological and structural organization using a user-compiled corpus containing 593 numbered records. The source material includes short phrases, sentences, repeated formulas, interrogative and conditional constructions, cumulative texts, pronunciation exercises, proper names, technical expressions, and extended performance materials. Because the corpus also contains exact duplicates, near-duplicate variants, source annotations, non-English entries, and records comprising more than one linguistic unit, normalization was treated as an essential methodological stage. The study employed qualitative corpus analysis to identify recurrent segmental patterns, feature-based phoneme contrasts, consonant clusters, syllabic positions, cross-word transitions, stress organization, rhythmic structures, lexical recurrence, morphological variation, and syntactic templates. The results reveal that English tongue-twisters are organized around several interacting mechanisms: repetition of a dominant consonant, alternation of phonologically similar segments, concentration of fricatives, recurrence of complex onsets and codas, maintenance of parallel syllabic positions, repeated lexical and morphological forms, constituent reordering, and progressive structural expansion. The findings indicate that tongue-twisting difficulty cannot be explained by phoneme frequency alone. It emerges from the interaction of segmental similarity, articulatory reconfiguration, syllable-position correspondence, prosodic timing, lexical overlap, and grammatical sequencing. A multidimensional classification is proposed for the subsequent annotation and quantitative analysis of English tongue-twister corpora.