We show how to construct a universal set of quantum logic gates using control over exchange interactions and single-and two-spin measurements only. Single-spin unitary operations are teleported instead of being executed directly, thus eliminating a major difficulty in the construction of several of the most promising proposals for solid-state quantum computation, such as spin-coupled quantum dots, donor-atom nuclear spins in silicon, and electrons on helium. Contrary to previous proposals dealing with this difficulty, our scheme requires no encoding redundancy. We also discuss an application to superconducting phase qubits.Quantum computers (QCs) hold great promise for inherently faster computation than is possible on their classical counterparts, but so far progress in building a largescale QC has been slow. An essential requirement is that a QC should be capable of performing "universal quantum computation" (UQC). I.e., it should be capable of computing, to arbitrary accuracy, any computable function, using a spatially local and polynomial set of logic gates. One of the chief obstacles in constructing large scale QCs is the seemingly innocuous, but in reality very daunting set of requirements that must be met for universality, according to the standard circuit model [1]: (1) preparation of a fiducial initial state (initialization), (2) a set of single and two-qubit unitary transformations generating the group of all unitary transformations on the Hilbert space of the QC (computation), and (3) single-qubit measurements (read-out ). Since initialization can often be performed through measurements, requirements (1) and (3) do not necessarily imply different experimental procedures and contraints. Until recently it was thought that computation is irreducible to measurements, so that requirement (2), a set of unitary transformations, would appear to be an essential component of UQC. However, unitary transformations are sometimes very challenging to perform. Two important examples are the exceedingly small photon-photon interaction that was thought to preclude linear optics QCs, and the difficult to execute single-spin gates in certain solid state QC proposals, such as quantum dots [2,3] and donor atom nuclear spins in silicon [4,5]. The problem with single-spin unitary gates is that they impose difficult demands on g-factor engineering of heterostructure materials, and require strong and inhomogeneous magnetic fields or microwave manipulations of spins, that are often slow and may cause device heating. In the case of exchange Hamiltonians, a possible solution was recently proposed in terms of qubits that are encoded into the states of two or more spins, whence the exchange interaction alone is sufficient to construct a set of universal gates [6,7,8,9,10,11,12,13,14,15,16,17,18,19] (the "encoded universality" approach). In the linear optics case, it was shown that photon-photon interactions can be induced indirectly via gate teleportation [20]. This idea has its origins in earlier work on fault-tolerant construction...