I see that another forum has finally established something that a 2-second analysis of the front door photos made utterly obvious (and which was posted in this thread quite some time ago):
a) The front door's lever-handle-operated catch was deliberately disabled (by wedging the catch open), since otherwise it would be impossible to open the door from the outside without assistance from the inside. This is because there was a lever handle on the inside face of the door, but (inexplicably) not on the outside face.
b) The door therefore needed to be locked by a key in the lower cylinder lock in order to keep it shut. Furthermore, once the door was locked in this way, a key was needed to unlock & open it from either the inside or the outside.
And, as Stilicho presciently said himself: "I think we can all see some problems in the prosecution scenario if the front door needs to be unlocked with a key from the inside"
I'm pretty sure that you didn't really mean your item "a)" to read the way it does. At least I hope not.
As I mentioned earlier, I expect that the lock assembly on that door is designed in such a fashion that the exterior deadbolt cylinder operates both the deadbolt
and the springbolt. This configuration is a 'key entry only' approach which is far from unusual. Where it was not well thought out is the idea of having a spring latched component which
mandates 'key entry only' from one side on a residential entry door. I expect that this was in a misguided attempt to design a system which would allow a resident to conveniently open the door while at home, say for visitors or delivery, without having to use their key each time, and without having to sacrifice security by having a door unsecured from the outside. I expect it was also compounded by the sales attraction of the unusual appearance. This was a trade-off in appearance and security v. convenience which backfired as many such trade-offs do, because of a user's insistence on convenience. The problem was compounded by the double cylinder deadbolt.
'Key entry only' in some fashion or other is very common on entry doors. I say "in some fashion" because alternatives are usually provided which suffer the handicap of being less secure under some usages, and the choice is left up to the user.
By way of example I am looking at the entry door to my apartment. It is a fairly typical configuration, and standardized in the hundreds of apartments in this complex. As is common with residential doors in the U.S. there are two"orb" type lock assemblies. One is a deadbolt, in this case a single cylinder deadbolt with an interior thumbturn. (Double cylinder deadbolts are problematic in such applications, not least because of fire safety issues, but also because there is no real increase in security. Once a perp has gotten inside a home there aren't a lot of good reasons to try and keep him there. The extra key cylinder doesn't make the lock any more tamper proof.)
Anyway, the other orb set in this instance is a simple "passage set", such as you might find on a closet door. It cannot be locked from
either side. It operates the springbolt ... and that is
all it does.
This configuration is elegant in its simplicity. Springbolt latches are inherently less secure (someone already mentioned the old credit card trick) and to have it lockable from the inside is essentially redundant, because the deadbolt provides the only real security available, and the thumbturn for it is right there, convenient. When leaving the apartment a key is required to lock the door, and thus there is no danger of forgetting your key.
The door hardware configuration in the Knox case seems to be an example both of design and marketing catering to misguided preconceptions (in the sense that "more is better") and of a failure to have thought out the inevitable consequences of the user's demand for convenience. The springbolt assembly, being inoperable from the outside, provided a certain sense of security for when the residents were at home and the deadbolt was not engaged. The deadbolt, due to its double cylinder application, seemed somehow "safer", even though that was illusory, but because of the inconvenience would only be used for such things as 'locking up for the night'. This was all just dandy, and plausible on the surface of it, until someone went to check the mail or visit the kids downstairs ... and locked themselves out a few times.
User ingenuity ensued. The spring latch was rendered inoperable by the age-old method of jamming something in it to keep it retracted. It is not needful to attribute fault to the owner, the installer, or even the manufacturer (whose product was not used as intended.)
So it really isn't "inexplicable" that there was no handle operating the spring latch on the outside of the door. But it is an interesting example of how door security can be over-thought and over-sold, how aesthetics can interfere with function in surprising ways, and why double deadbolts on residential entry doors are more of a PITA than of any real advantage. I would be surprised (not certain, though, which was why I was trying to find specs on this particular mortise assembly) if the lock body itself wasn't designed to accept an exterior handle, and that one could be retrofitted with comparative ease. (For a door hardware installer, at any rate.) Changing the interior deadbolt cylinder to a thumbturn is child's play.
--------------------------
More interesting to me is thinking about the
consequences of defeating that spring latch in such a fashion. In all likelihood the result of jamming that latch back would be apparent to anyone trying to operate the lever handle on the inside. I admit that I would be more sensitive to such discrepancies than most people because of my experience, but this probably wouldn't have been subtle. The handle would have felt, well ... floppy, I guess. Or maybe rigid and stuck. (All this depending on the working mechanism of the lock design.)
Nearly anyone who tried to use it would have immediately thought, "This thing ain't right." It would catch their attention. If they had any thought of securing the door for whatever reason, harmless or malign, they would have been drawn to pay extra attention to it.
I stress this because the suggestion that Guede would have locked the entry door when leaving has been so cavalierly dismissed by the idea that "he thought it was latched. I just don't think so.
I was skeptical before when all I had to go on was a nebulous assertion that the latch was "broken", but now than I have seen exactly what "broken" really consists of I am more than skeptical.
I can imagine as many scenarios as anyone else that rationalize why Guede would have so purposefully locked the bedroom door, but so thoughtlessly left the entry door unsecured. Maybe even more. Most are possible, some even plausible (sort of), but none are probable or likely.
What is quite likely is that someone who was attempting to fabricate the impression of a stranger intruder, and having to support that fiction over different interviews at diferent points in time might not have thought out the inconsistencies implied by the state of the two different doors. The relatively unique configuration of this door and its "user modification" created an inconsistency which needed to be glossed over. To me it is another stone in the wall.
I think Knox initially claimed that she found the entry door unsecured because in her mind, at that moment, it somehow reinforced the stranger intruder scenario. I think that Meredith's door was locked in an attempt to account for her failure to glance in the room even after having "discovered" so many other signs that something was wrong.
I have to think that in retrospect she wishes that she hadn't done one or the other. Nothing about the state of the doors could ever constitute acceptable evidence in a courtroom, but I have no doubt that it set off alarms in the mind of any competent investigator. It certainly caught my attention.