Case One - Vacuum: The water molecules vaporize almost instantly. At 25°, one square meter of water surface will vaporize some 3.41 kilograms of water per second in the absence of any surrounding gas. That is equivalent to a column erosion of some 3.42 millimeters per second, and a numerical vaporization rate of some 1.14 x 1026 molecules per square meter per second.
Case Two - Air at Equilibrium Vapor Pressure: Firstly, let us replace your intermolecular ropes with fairly rigid springs. The angles at which water molecules form their intermolecular hydrogen bonds have preferred values. Any deviations from these values require the application of force. Secondly, water—being a fluid—has no rigid structure. The molecules cannot and will not keep their relative positions. Thirdly, hydrogen bonding is ephemeral. At 25°C, the average liquid water molecule breaks all of its hydrogen bonds with its neighboring molecules and forms new bonds with new neighbors many billions of times each second. Even a surface water molecule making up part of the surface tension network will vaporize and be replaced some ninety billion times a second. At equilibrium vapor pressure, the number of new arrivals and the number of escapees roughly balance. Fourthly, molecules are in random movement. At rest there are just as many water molecules moving in anyone direction as in any other direction.
As the water falls, more molecules will have a downward component of motion than in any other direction. Since pressure is the simple product of number of impacts per unit area and time and the mean impulse per impact, this reduction in lateral motions is reflected in the diminution of lateral water pressure (the Bernoulli Effect). Meanwhile, the air pressure remains the same. The consequence is increased relative lateral pressure on the water column and a diminished diameter.
Arashmh, did I give you the molecular explanation you were looking for?