1: \begin{abstract}

2: % Background

3: The Neumann problem of linear elasticity is singular with a kernel formed by the 

4: rigid motions of the body. There are several tricks that are commonly used to obtain a non-singular linear system. However, they often cause reduced accuracy or 

5: lead to poor convergence of the iterative solvers.

6: %The singularity can be addressed in several different ways, but to construct 

7: %iterative methods that yield convergent numerical solutions in a bounded number of 

8: %iterations, the kernel must be properly treated.

9: % Methods

10: In this paper, \deleted{four} different well-posed formulations of the problem are studied 

11: through discretization by the finite element method, and preconditioning strategies 

12: based on operator preconditioning are discussed.

13: % Results

14: For each formulation we derive 

15: % \replaced[id=rev2, remark={comment 1}]{formulation}{problem}

16: preconditioners that are independent of the discretization parameter. Preconditioners that 

17:   are robust with respect to the first 

18: Lam\'{e} constant are constructed for the pure displacement formulations, while 

19: a preconditioner that is robust in both Lam\'{e} constants is constructed for the 

20: mixed formulation.

21: % Conclusions

22: It is shown that, for convergence in the first Sobolev norm, it is crucial to 

23: respect the orthogonality constraint derived from the continuous problem. Based on 

24: this observation a modification to the conjugate gradient method is proposed that

25: achieves optimal error convergence of the computed solution.

26: \end{abstract}