With the possibility that they may be neutralised in Syria, now is perhaps a good time to ask why chemical weapons have not been used more extensively in warfare. The question is often asked in the context of WW2, given the precedent of mustard gas in WW1 and the fear that led to the general distribution of gas-masks at the start of the later conflict. Popular answers include a willingness on all sides to observe the 1925 Geneva protocol banning such weapons (which may be confusing cause and effect) and Hitler's own experience of being gassed in the trenches (which didn't lead him to outlaw Zyklon-B). In fact, if you consider the second world war to have started with the earliest aggression by Axis powers, then chemical weapons were used by Italy in Abyssinia and Japan in China in the late 1930s. The eurocentric view is rather blinkered.
The real reason for their limited use is that chemical weapons, particularly airborne gases, are of little military value due to their lack of discrimination and can even be counter-productive, as the tales of a change in wind causing blowback in WW1 attest. The stalemate of the trenches also explains why they were used there. The hope was that they could provide a more effective means of clearing the line ahead of another big push, shelling having proved ineffective due to digging-in, but the key enabling factor was proximity, in much the same way that distinguished pre-industrial chemical weapons, such as catapulting a disease-ridden animal corpse into a besieged town or poisoning an accessible water supply. Without the trenches of the Western Front, it's unlikely that gas would have been used.
There was limited tactical use of chemical weapons during WW2, notably the defoliant Napalm in the Pacific, but they were never decisive on the battlefield. Even as "terror weapons", chemical agents would not have been as deadly as the incendiaries that created firestorms at Coventry and Dresden, or as psychologically stressful as the random V1 and V2 rockets that hit London. Though the debate around the strategic value of mass-bombing continues, and in particular the balance between hitting military targets and hitting civilians, it is clear that no one thought that terrorising the population alone would win the war, or at least not until Hiroshima and Nagasaki.
Ultimately, the unwillingness of the belligerents to deploy chemical weapons during WW2 (both sides had large stockpiles of mustard gas) was probably down to a combination of their poor military utility and the fear of reprisals. Since then, they have been used occasionally, mainly in the context of civil wars: Yemen in the 60s, Vietnam and Angola in the 70s, and possibly Afghanistan in the 70s as well. Iraq used chemical weapons, mainly mustard gas and sarin, against Iran in their 1980s war, but it's most infamous use was against Iraqi Kurds at Halabja in 1988.
It is claimed that chemical weapons are "too horrible, even in war", but the suggestion that there is an ethical red line is nonsense. There are plenty of conventional weapons just as horrible, and they kill people just as effectively, while the ultimate deterrence of nuclear weapons is premised on a willingness to kill more people in a hour than have died in all the wars throughout history. Some part of the visceral distaste for chemical weapons stems from the use of gas in the Nazi death camps, and there is much truth in the observation that the nuclear powers don't want lesser states to access "terror weapons on the cheap", but another reason is the realisation that while they are often ineffective on the battlefield, they are well suited for use against an adjacent civilian population. In other words, if you want to kill indiscriminately, and aren't particularly concerned about the risk of collateral civilian losses on your own side, then chemical weapons may fit the bill. This is the practical lesson of Halabja.
In this sense, Assad's recent claim, that US intervention in Syria could cause the whole region to "explode", is not mere bluster. The use of chemical weapons, assuming Assad's forces were responsible, looks like a demonstration, a threat of wider chaos if the West directly threatens the regime. The red line then is not about the use of chemicals weapons per se, but about the escalation of the Syrian conflict and its potential spread to neighbouring states (the metaphor of an indiscriminate gas cloud at the mercy of the winds is obvious). The dilemma for the US is that intervention might just as effectively provoke this very result, which is why a face-saving exit strategy has been engineered this week (the idea that John Kerry's suggestion was a gaffe is absurd - diplomats do not make unplanned remarks in press conferences).
I've mentioned previously that the US's policy in Iraq can best be explained by assuming that the current state, a fractious and emasculated power, was the desired outcome. In this sense, it's worth noting that the developments of the last 3 years have similarly undermined Syria as a regional power, and at a time when Egypt has been forced to the sidelines by domestic upheavals and Turkey has begun to raise its profile as a regional player. I doubt the US has a policy of deliberately "balkanising" the Middle East, but it does seem comfortable with the emerging patchwork of sectarian enclaves, just as it remains comfortable with the patchwork of Gulf monarchies and emirates.
It's worth remembering that the bogey for the West (and Israel) in the region has long been pan-Arabism, the idea that the post-Ottoman era should have led to a unified Arab state stretching from Egypt to Iraq (there was a short-lived union between Egypt and Syria between 1958-61). The chief impediment to that has been the tendency of various regimes to manipulate and exacerbate religious, ethnic and tribal divisions, paying only lip-service to the idea of an Arab nation. The willingness to use chemical weapons against your own people is as much a symbol of that as the scapegoating of Copts in Egypt or the pitting of Sunni against Shia in Iraq. The fear is that whatever the outcome of the current decommissioning initiative in Syria, chemical weapons remain more likely to be used in the Middle East than elsewhere.