OK, Maybe i can help clear this up!
First, the term "drive" as used in training working dogs is a way of grouping and naming behaviors so we can communicate about them for the purpose of training and propagating working dogs. Drives are not simply a trait that is inherited by simple recessive or dominance, or this would be easy, they'd either have inherited it or not, and they would have it as a phenotype or not.
But, when we view groups of behaviors and name them drives there becomes such a great variability in individuals it defies being quantified.
So, the important thing here is to remember that no drive exists, we just name particular groupings of behaviors as "drives".
Then we have to look at the definition of those groupings. Now don't get too worked up over the dogs evolution and why they once did things when and if they were once wolves. Domestication has taken the village mongrel, kind of a cockroach of the canines, and created sooo many variations that getting hung up on wolf behaviors isn't helpful.
The definitions are subjective although based on observation, they are the result of many many years of observations and are handed down to us from one german state's police dog program. Much of their terminolgy can be traced back to behavioralists like Lorenz when the term "drive" was still in use by behavioralists (now among researchers it is deemed an antiquated term, not useful in study since it really doesn't discribe the behavior you are looking at adaquetly to be quantified).
Now, among many in the police world, there is an educational process to help people involved in the vocation of police dog training to evaluate and train for the particular standards they seek.
To do this they have to be on the same page and that is where the use of drive as a term and in defining behaviors comes in. If I say to someone, that dog is functioning in fight drive it does not exclude other drives, it cannot because it is simply a defining term for the emphasis we interprete the dog to have in its work.
To be sure, we train together and look at a dog and how they react and compare notes. Until the internet this system was great. Now we cybertrain and a flat screen cannot, and will not, be able to provide the vast amount of information the dogs give us while they work.
Now the question of fight drive.
Well, first lets define what it is not. It is not the desire to grab a substitute rabbit, like a ball, or a tug, or a sleeve. It is not the desire to make a threat go away by posturing, barking, growling, looking bigger, or by biting. Nor is it the education provided where the dog learns that it can bite from this feeling of concern for himself and turn it into a comfortable feeling of holding and carrying a prey item by biting. This is a common misconception, that somehow when these are put together I have fight drive. It can look this way to the novice but it is not. Now while you get up in arms do not forget that many many dogs have some degree of fight drive and that this system does not negate fight drive. Here is where the confusion exists. Anyone in dogs can see that among the dogs trained in this classic way there are powerful fight drive dogs that exist.
It can also be seen that dogs trained to treat the sleeve as a big rabbit can also display fight drive, they just cannot NOT see the antogonist making it go.
Since, we see it in conjunction with many training approaches it is often seen as associated with the method and not the dog.
So, fight drive, which is most frequently brought into the picture among dogs being trained in classical ways and not often existing alone is confused as an interaction between prey and self defense. The truth being that it can, though seldom does, exist as a group of behaviors, that are desireable among the working dog because it is the only way that we can have a strong confident working dog that behaves in the following fashion: the dog fights when fought and escalates the fight based on its opponents level of combat. If the opponent becomes passive the dog has won, if the opponent retreats, the dog has won, if the opponent fights the dog fights.
In a prey centered dog that has little fight to fall back on, if the prey is pursued but turns to its pursuer prey drive says "hey, this prey is dangerous and too costly to grasp, time to back off" it may even be brought into defense where it would back up and try to scare the prey (now also acting in self defense) through posturing and barking and if backed into a corner even biting. Conversely, if the prey is defeatable and weakening between attempts at escape the dog intensifies its prey behaviors and its confidence rises (producing big grips in foundational training is based on this).
The difricult to understand part of this work comes from peoples' inability to understand that identifiying a threat and being afraid are not necessarily linked all the time.
I explain this to the people I train that are Law enforcement, that this is like making a tactical approach to an alarm, or traffic stop, you are not afraid but you are dealing with a threat. It is often that people equate threat with fear.
Now, the argument is made that this had to start due to a fear response or how could they have identified the threat? This is a little simplistic but, it is hard to imagine an animal who never ever became afraid of something, so this becomes a circular argument. The bottom line is that a dog doesn't evaluate the moment based on an occasional event in its life, that is why training takes so many repetitions in all areas, it responds based on some set of inate behaviors we select for and we capitalize on. In this instance we name them fight drive.