Please translate the following to simple English:
In order to improve the readability of this section, all numbers mentioned refer to the frequency of some morpheme tag in the spoken corpus relative to the written corpus. For example, a relative fre- quency of 10 means that the tag in question ap- peared 10 times more frequently in the spoken cor- pus than in the written corpus.
The first result to note is that of verbs and copula- tives. While the spoken corpus contains 4 362 ut- terances, the written corpus contains 712, which is a ratio of about 6 spoken utterances to every writ- ten sentence. However, two typical kinds of verb phrases, namely verb based and copulative based verb phrases, occur only 2 and 3 times as often in the spoken corpus. This is surprisingly low, and seems to indicate that the utterances in the spoken corpus tend to lack verb phrases. This may be because of in- terruptions that occur during a dialogue, or it may be some form of ellipsis.
A feature that stands out, however, is the impera- tive, as suggested by the relative frequencies of the imperative prefix (about 10) and imperative suffix (almost 7). This is consistent with the summary provided by Akinnaso (1982), who mentions imper- atives alongside interrogatives. In our experiment, both interrogative tags in the ZulMorph tagset had a relative frequency of about 4. This is especially in- tuitive considering the nature of the spoken corpus, which typically takes the form of a dialogue between characters in a soap opera. It is therefore also unsur- prising that the relative frequency of the first person singular morpheme tag is 7.5, while the second per- son singular tag has a relative frequency of almost 3. We note that the first and second person plural tags have significantly lower relative frequencies, namely 1.5 and 0.9, respectively. In fact, the second person plural is one of only two features to have relative fre- quencies below 1, indicating that the feature occurs more frequently in the written corpus. However, in this case, the number is very close to 1, and therefore rather indicates that the feature occurs equally fre- quently in both corpora.
The other feature occurring more frequently in the written corpus is the passive voice, which again ac- cords with the literature for English. Here, the pas- sive voice is almost twice as frequent in the written corpus as in the spoken corpus.
We note that the negative prefix has a relative fre- quency of about 2.5, consistent with the literature for English. isiZulu does not have an explicit defi- nite or indefinite article, but demonstratives have a
5
relative frequency of about 2. Cornbleet & Carter (2001) state that “various differences” can be found between written and spoken English with regards to time relationships, and in this work we see that es- pecially the use of the future tense is more frequent in the spoken corpus, although the past tense also occurs slightly more frequently.
One instance where a clear confirmation was not found was in the case of gerunds, which we approx- imated for isiZulu by counting noun stems from class 15, the class of infinitive nouns (Poulos & Msi- mang 1998). Contrary to gerunds in English, the spoken isiZulu corpus did not exhibit fewer infini- tive nouns than the written corpus. This is likely due to the fact that infinitive nouns in isiZulu are not sufficiently equivalent to gerunds in English: in- deed, infinitive nouns have a “dual nature” (Pou- los & Msimang 1998), and a more syntactically in- formed investigation would be required to differ- entiate their nominal and verbal usage in the two modalities.
Our investigation has shown a basic similarity be- tween isiZulu and more well-studied languages, such as English, for features that can be identi- fied morphologically. The similarities found on the morphological level would suggest that other rel- ative differences between spoken and written lan- guage at the syntactic and semantic levels, may also be exhibited by isiZulu.
To make this section easier to understand, we look at how often certain word tags are used in spoken language compared to written language. For example, a relative frequency of 10 means the tag appeared 10 times more in spoken language than in written language.
One notable finding is that verbs and copulas occur less frequently in spoken language compared to written language. Imperatives, however, are used more often in spoken language. Interrogatives also have a relatively higher frequency in spoken language.
The passive voice is more common in written language, while future tense and past tense are slightly more common in spoken language.
Gerunds in isiZulu are different from gerunds in English, as infinitive nouns in isiZulu have a dual nature. Similarities between isiZulu and English are found on a morphological level, suggesting there may be other differences in syntax and meaning between spoken and written language in isiZulu.