Yes, if two words are distinct only by a single vowel – especially an unstressed one – the are very similar, and might easily get confused in a noisy environment. The same is true, for example, with pairs of words like [neʧi] and [neʨi] – that differ only at one point in the shape of the tongue. Whether two sounds are similar or "very different" depends on your linguistic background. Some people might find unstressed and [o] very similar, and might tend either to merge them or mix them up. Others might fund various post-alveolar fricatives difficult to tell apart, etc.Rodiniye wrote:
Different vowels is similar sounds to you? Especially when we are talking about vowels with very different sounds?
It's not necessarily a problem that languages have lots of similar-sounding words. In fact, natural languages often have lots of identically-sounding words – homophones – and this is rarely a problem. The fact is, we don't grasp words by hearing the phonemes one by one, and then putting them together. Rather, we fill in quite a lot from the surrounding context – both the linguistic context, and the extra-linguistic (pragmatic) context.
That the same word might belong to several different parts of speech, need not be any problem. Even if there is no overt change in the form of the word. This is frequently the case in English, where the same word can be a noun and/or a verb and/or an adjective and/or possibly also an adverb or something else. This is usually not a problem because the syntactic behaviour of nouns, adjectives and verbs are usually very different, and one can often easily tell from the syntactic environment whether a word is a noun, verb, adjective or something else. There hardly any need to make use of Esperanto-ish rules like "every noun ends in this vowel, every verb ends in that vowel". One could of course, if one liked to, but the absence of such rules does hardly pose any problems.
Agreement patterns may also help to tell near-identical words apart. Word with different genders might be distinguished only by a single letter. Or they may even be identical. But they trigger different forms of agreement – accompanying adjectives, articles, determiners, verbs, etc. may take different forms depending on the gender or the nouns, and this helps us to understand the utterance correctly.
And finally, the general, non-linguistic context may of course frequently be of great importance. If two similar-sounding – or even homophonic – words are likely to be used in very different contexts, the risk of confusion is small. If, on the other hand, two near-identical words are likely to be used in the same context, the risk for confusion is greater. Say that, in language X, the same verb root could be used to form nouns indicating agent, instrument, patient or result etc. of an action, by simply changing a single sound at the end. Then there might be a risk for confusion. Of course, the only way to know for sure would be to foster a community of people speaking language X, and see whether they would often confuse the derived nouns. Maybe they would, maybe they would not.
I'm not against the notion of an international auxlang per se, though I'm critical of the naïveté that seems to lie behind many auxlang projects. I really hope you understand that 20 pages is nowhere near a "complete grammar". There's a *lot* more needed. (And as I look through the last few pages, I can see that there are still questions that hasn't been answered.)