Symbols:- The wavy line represents the external load on the bridge due to traffic and pedestrians.
The full arrows are the external force reactions from the bridge foundations. The dashed arrows are the internal forces within the bridge.

Explanation:- The downward external (wavy) load is resisted by an upward axial tension in the cables.
However because the cables are at an angle the tension in them also compresses the deck.

The design concept by Calatrava was apparently based on the following reasoning.
1. The cables pull on the tower which resists, not through back stays as would normally be the case, but by the sheer weight of the tower.
2. The cable tensions and the weight of the tower combine to create an axial compression along the axis of the tower.
3. The downward axial compression in the tower pushes horizontally against the bridge deck.
4. The compression from the tower balances the axial compression in the deck.
4. As a consequence the net force on the foundation under the tower is vertical.

Unfortunately this can only work for one set of loads on the bridge.
In reality the loads change and indeed the construction tolerances in the actual weight of the tower are such that the forces rarely balance precisely.
Indeed the bridge is very sensitive to small changes in geometry or weight because its structural configuration relies on setting to zero the difference between two large bending moments.
So maintaining only an axial force in the tower would require the weight and inclination of the tower to change too - something that is clearly not feasible.

When all loadings and variations were considered by the construction designers and advisors and when they calculated the envelope of the possible maximum and minimum internal forces in the bridge they realised that they had to design bridge deck, the tower and the foundations to cope with very large bending moments.
They realised that small variations in the weight of the tower and the forces in the cable stays could affect the safety of the bridge
Consequently adopted a sophisticated quality control system to check the weights as built and to measure the forces in the cables. They used the actual measured values in a computer programme to check against the performance of the actual bridge so they could be sure they understood how the bridge was behaving as it was built.
They realised that they must also allow for creep and shrinkage of the concrete in the tower.
A huge foundation had to be built with a large number of very large piles of big dimensions and a huge capping plate on top to connect them.
Direct evidence of the structural inefficiency of this type of bridge is that the maximum bending moment in the deck is around 60 % of the bending moment of a simply supported beam with the same span.