You're so close Krauss to making the good argument. So close. You were almost using the (good) terminology of logical positivists. You asked "what does it mean" (or something close). You almost got to asking "You argue that there can not be objective morality without god. What would a world without god look like? How would the world be different if there was no objective morality? Would it look different? How could you tell which world you were in?
" There is no answer to that question. That's why the term "objective morality" as used by Craig is simply meaningless.
In other words, Craig's argument relies on a conflation of the several meanings of "realism", "objective", and "absolute".
"Realism" is defined only in the context of science. "Realism" itself is rather ill-defined. We can talk about how things might be different, and what observable differences that might make, but that's as far as "realism" is meaningful.
"Objective" can mean that the truth of an objective statement is derivable from the truth of other, more basic, statements.
"Objective" can mean that the values and process leads to the same conclusion for every reasonable practitioner. Judging football is objective. Judging figure skating is not. Science is objective.
"Objective" can be a synonym for "realism".
"Absolute" can describe a measure. An absolute measuring system is one in which there is an unambiguous "better" or "greater". Perhaps it implies a total ordering, or perhaps it implies only a partial ordering. http://en.wikipedia.org/wiki/Total_orderhttp://en.wikipedia.org/wiki/Partially_ordered_set
For example, someone who is a cultural relativist does not have an absolute morality.
"Absolute" can be a synonym for any of the above senses of "objective" and "realism".
Each of those different sense are important, and they are not the same. It is coherent to talk about a system with a "greater than" comparison which is derived from more basic values, and another which is not derived from more basic values. It is coherent to talk about such systems which can be used to come to universal agreement like judging football, and those which have ambiguity, such as judging figure skating. We can talk about systems with a "greater than" based on (scientific) realism, such as "longer than", and those not based on scientific realism, such as any morality with a "better than" comparison.
The morality of Craig is not realism, despite his protestations to the contrary. It is objective in the sense that it can be derived from more basic values, but only if you allow the silly option of breaking Hume's is-ought distinction. It is not objective in the sense of realism. It is objective in the sense that the values and process lead to the same conclusion for everyone who adopts those values and uses that process. It is not absolute in the sense of realism. It is absolute in the sense that we can say that some things are better than other things.
My morality of humanism is not realism, nor objective in the sense of realism, nor absolute in the sense of realism. My morality is not objective in the sense that I derive it from other, more basic, value or statements. (I do not break Hume's is-ought distinction.) It Is objective in the sense that every practitioner arrives at the same conclusion, and it is absolute in the sense that some things are better than others under the system.