Like some others on this thread, I'm not a Christian and have no personal investment in the Trinity doctrine. I've studied the concept carefully though, and have come to the conclusion that there was an evolution of the concept of the nature of Jesus. The earliest Gospels contain hardly a trace of an idea that Jesus and the Father were part of a Godhead, but this concept seems much more developed in the Gospel of John, although it's still not directly stated. Over many more years, non-canonical Christian writers drifted more and more towards various forms of the Trinity.
As for John 1:1, most trinitarians simply have no idea what they're talking about. An excellent book, Jesus as God (Murray Harris, early 1990s), goes into gruesome detail on John 1:1 and a few other key texts that trinitarians find critical to their belief. Harris shows that theos in John 1:1c has a qualitative aspect, so that the base Greek "kai theos en ho logos" is literally "and god was the word", with "god" having the meaning of not God himself, but the nature of God, whatever that was. Harris offers several alternatives that more closely match the Greek meaning, such as "what God was, the Word was" or "the Word had the same nature as God".
I don't find these ideas particularly satisfactory, though, because they really beg the question of what John had in mind by saying something like "the word had the nature of god." What exactly is the nature of "god"? Did John have in mind the nature of the God of the Jews, Yahweh, in the sense of his nature as The Supreme Creator and Ruler of the Universe? Or did he have in mind the generic meaning of "god", which certainly includes the pagan gods, the devil and even powerful humans in everyday koine Greek use?
Let me illustrate this with an example. Suppose there's a group of intelligent monkeys. A human finds them and enslaves them. This man is the only human the monkeys have ever encountered, but they know that other humans exist. The man sets himself up as an all-powerful ruler over them, and becomes, in effect, their God. The monkeys come to call the man The Human. The Human also tells them his name is George. Now, suppose another man named Jack comes along and interacts with the monkeys for awhile, and then leaves. Here's my point: if a monkey writes "human is Jack", could one tell only from the writing that he meant that Jack is a member of the category called "human", or did he mean that Jack is a member of the category (with only one member, who is named George) called "The Human"? The answer is that you simply can't tell.
It's the same thing with John 1:1c. Murray Harris argues (poorly, in my opinion, and in line with other scholarly trinitarian thinking) that John's use of theos must have been restricted to mean God, and only God. But in normal Greek usage, theos was an extremely broad category with millions of individual members, just as human is a broad category with billions of individual members. To simply declare what John meant, based only on textual considerations, is to make an unfounded claim.
Just as one can properly write, "human was Jack", one can also properly write, "god was the word". One can also properly write, with exactly the same meanings, "Jack was a human" and "the word was a god". So in this sense, the New World Translation is correct in its rendering of John 1:1c. And reference to the KIT doesn't do naysayers any good here.
The point is that one cannot properly use the Greek text of John 1:1 to argue for or against the trinity. The text is neutral in this regard, and most careful scholars, when pinned down on this, will agree, just as Murray Harris -- an ardent trinitarian -- tacitly admits in his book (he argues that the NWT rendering of John 1:1c is wrong on grounds other than textual considerations, but I think his arguments are weak). In fact, Harris explains clearly why the standard rendering, "the Word was God" is flat-out wrong, because that phrasing implicitly equates God and the Word -- which is a heresy according to most modern Christians. That's why he argues for other renderings which are more precise in giving the original Greek meaning. However, as a trinitarian scholar with an audience he has to play to, or lose them, he can't admit that the NWT rendering is allowable.
AlanF