As I explained in the draft of my latest paper about linguistics, I have measured the collision entropies of English, German and Croatian syntax. The entropy of consonants in a long English text was log2(11), whereas, in an English spell-checking word-list, it was log2(13). Therefore, the entropy of the English syntax was log2(13)-log2(11). For German, the entropy of the syntax was log2(15)-log2(12). But log2(15)-log2(12)>log2(13)-log2(11). And, I hope we agree, English has a more complicated syntax, as English syntax is often requiring meaningless words such as "the", "that", "than", "it" (dummy pronoun)... So, how is this possible?

  • 1
    Make sure that your presuppositions on German and English syntax hold for the corpora you have chosen. Try other measures of syntactic complexity and look how they compare. Commented Oct 12, 2021 at 17:25
  • 2
    How exactly does German syntax not also require these exact same types of “meaningless words” (by which I assume you mean function words)? Commented Oct 13, 2021 at 23:43


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.