gypsy,
You are really in a certain mode now aren't you. It's difficult to respond to every point in your lengthy point, but I will try to clear up some of the questions by presenting my understanding.
First let's define what we are talking about. When we speak of and criticize Lydiard, it's usually something like the classic representation you can find at the Lydiard Foundation website:
http://www.lydiardfoundation.org/pdfs/al_training_eng.pdf
. On one of the pages, you will see an outline something like:
1) Marathon conditioning - as many as you can (depends on previous season and the next race)
2) Hill training - 4 weeks
3) Anaerobic training - 4 weeks
4) Coordination - 4 weeks
5) Freshening up - 2 weeks
6) Continuation of Racing
Other books and representations give different period lengths and outlines (they are all supposed to be examples of balanced implementations), and the same presentation also mentions "complex" examples like the continuation of racing and the race weeks/non-race weeks, but for the purpose of explaining the comment complaints, I will assume the summary above.
My understanding of "complex", for the purpose of my explanation, simply means that all the elements of training are represented and trained simultaneously, within a regular cycle. For example, everything will be trained at least once, on a weekly basis (or maybe a 10 day cycle). This contrasts with the classic periodization, where each phase has a focus on developing a single training variable (and maintaining previously developed variables). When Antonio speaks of linear versus non-linear periodization, this is the distinction I understand. Lydiard's classic periodization is linear, while a non-linear periodization is complex.
Lydiard presents continuation of racing, and the race week/non-race weeks. Robert de Castella also used a "complex" implementation of Lydiard's ideas. So these are alternate examples of "complex" training linked to Lydiard ideas. Although they are called complex, they can be easily described with "typical week" schedules. But I think these are not good examples of what we mean by "modern training" with non-linear periodization -- these typical weeks are not many periods, but basically the last period before the cycle starts again.
When you read some of Canova's threads, describing some of his theoretical ideas, and some practical examples, these are not simple plans easily explained by "typical weeks". In fact, like Lydiard, he also describes different periods. At first glance, it's hard to understand the difference. I don't want to summarize it here, but a lot of the information can be found by reading the first few pages of this thread:
http://www.letsrun.com/forum/flat_read.php?thread=2959804
(Note his name changes in the thread -- I think there was a story when someone guessed his password, so Renato started posting as "Canova Renato").
A brief outline is that after a brief INTRODUCTORY period, the training consists of a FUNDAMENTAL, then SPECIAL, then SPECIFIC period. Each phase provides support for the next one. If you want more details, it's worth taking the time to read and parse those threads. There, the paces are better described (REGENERATION, GENERAL, SPECIAL, SPECIFIC), as well as the concepts of EXTENSIVE training, versus INTENSIVE training. The approach is also different for middle distance, long distance, and marathon training. But that's a pretty good thread for presenting most of the concepts, although if I recall, it doesn't mention the use of CIRCUITS (like Lydiard's hill training), which were detailed in the long Canova thread.
I asked Antonio once (well more than once) to explain how we could "fix Lydiard" by going from Lydiard to modern training. He seemed reluctant to do it, as if I asked him to make oil from water, but I actually liked his answer. If you start with Lydiard phases, you would pull the interval training forward into the marathon conditioning phase (they would form a defined progression towards your event as the training advanced), and the hill training phase would be extracted, and spread across all phases (as some sort of strength training or resistance training circuits). He used a little more detail (describing the extensive and intensive training, and progression), but for some reason, the image of morphing from one to the other really helped put things in a better perspective.
So that was just the starting point of the definitions and concepts, as I understand them.
It's not correct to think that a modern approach neglects aerobic training. It is not relegated to any kind of secondary status. Using Canova's example, the FUNDAMENTAL phase provides the AEROBIC support. The main difference is that, under Lydiard, it is generic for 800m to marathon, while under Canova, the pace is qualified with respect to race pace. Aerobic support required for 1500m is not the same as aerobic support required for a 10K or a marathon. Another distinction (not found in the thread I linked) is that Canova already describes the use of circuits (different ones for FUNDAMENTAL, SPECIAL, and SPECIFIC periods), and will include interval training in this phase (at least for middle distance). These intervals will not have intense repetitions, but the goal will be to do more of them, as the training progresses, or reduce the recovery interval. So for Lydiard, everyone uses the same definition of AEROBIC, but for Canova, the definition of AEROBIC is refined, as it takes the athletes event requirements and ability into consideration.
It's also not correct to think that Lydiard completely ignores specificity. As I've argued before, Lydiard felt specificity was very important (once you were ready for it). He featured it as a reason why Murray Halberg won a key race, by pulling a 70 yard lead in one lap, because his training included a progression of race distance time trials and development races, while his competitors were only interval trained, and their interval training didn't prepare enough them for the actual race conditions. The main problem is that these kind of "specific" time trials and races don't really appear until the coordination phase (e.g. the 4th month of the training cycle). The earlier phases (marathon conditioning, hill, and anaerobic training), are not defined in terms of race pace.
So what did I mean when I previously mentioned "without too much negative physical consequence"? It boils down to this question. Is there a valid reason to do an extended phase of aerobic only training? After all, racing requires a combination of all the training variables, at the same time, during the race. Why doesn't it make sense to take a "holistic" approach, and train the variables simultaneously? Speaking for myself, after an intense season, I needed a break purely for mental reasons. So it was important to run, without aggressive goals, just to bring the fun back. But is there a physical consequence? For example, would a 3 month "aerobic only" phase lead to an inferior preparation, come race day, versus a modern "complex" approach? I was expressing my opinion that I don't think it ever hurt me, either in the short term or the long term. (Actually, when I competed, what hurt me was the bad training during the season, and not the training during the off-season, but that's another story). Another part of the same question, do many years of many cycles of "aerobic only" phases lead to a true lack of reaching one's potential? You might also ask the question from the other side: does it lengthen your career, or allow you to reach a higher potential? I think this is one of the main areas of disagreement. Does a linear periodization approach help you, hurt you, or is it immaterial.
Maybe to address Antonio's "aerobic" comment, while I address the VO2max benefits after 2 years point. I think here is also a misunderstanding. It's true that Lydiard says that aerobic development is quite unlimited. Antonio has expressed the fallacy of that belief many times. Antonio has also argued that interval training can bring a superior aerobic development, versus continuous efforts (e.g. a 52 minute 10 miler). I don't speak for Antonio, or Lydiard, but just setting the context for my opinion. I gave a reference that said VO2max changes occur fairly quickly, in the terms of one career. But just because "aerobic training" starts bringing less and less "aerobic adaptations", doesn't mean that other benefits stop too. I think there is (or should be) a "practice makes perfect" principle, that says benefits still come from just running. The more you run, the better you get at it, improving efficiency and economy. The oxygen cost of your efforts decrease, for the same pace, because of improvements in neuro-muscular systems, and reduction of wasted movements. This effect last on the scale of decades (if you are lucky). You don't need to give up the "aerobic training" completely, but having said that, I think it is most important in the early years, and becomes less important for the seasoned and mature athlete. Renato said something differently, that athletes first work on increasing their volume, and then later in the careers, the volume will decrease, while intensity increases. Also, to make the same point again, yes I think with children, it is important to build an aerobic foundation first, before starting serious training. And again, this is how I understand Renato when he says that Africans start their training already at 85-90% of their potential, since they are already doing what is the basis of training, running long and fast.
But I said before, I don't understand the value of talking about things like capillaries and mitochondria, even in the first two years, except maybe to explain it once to the beginning athlete to address a minor curiousity. We can measure improvements with a stopwatch.
I think I have to stop for now. I have to give Antonio credit for calling me out, after I said I wouldn't participate too much in this thread.