He appropriately raises some of the limitations of meta-analyses in gen-eral that we thoroughly address in the discussion, including their retrospective nature, and the possibility of heterogeneity in results, data availability bias, and publication bias.We concur that compared with the gold standard of randomized controlled trials, the results of meta-analysis need to beweighed critically. This is whywe performed careful scrutiny of articles with multiple sensitivity analyses to mini-mize the risk of these various biases influencing our results; while the inherent limitations of meta-analysis remain, we believe that the ro-bustness of our findings to multiple sensitivity analyses supports our overall conclusions. Unless large controlled trials are conducted to address these difficult questions, careful pooling of the available data will need to provide the best available answers. Our report is presently themost thoroughexistingdataanalysisonhighgradeT1 tumors, and

thoughtful comments and suggestions. 2

for thoughtful comments and suggestions.

Sayers and Flikweert, 2012 Understanding and communication confidence in flood risk information – A discussion note. 1 | P a g e

especially grateful for the thoughtful and thorough comments of the referee which greatly improved the paper. Geert Bekaert thanks the NSF for financial support.

Criticism invariably consists in pointing out some contradiction; either a contradiction within the theory criticized, or a contradiction between the theory and another theory which we have reason to accept, or a contradiction between theory and certain facts.... Criti-cism is, in a very important sense, the main motive force of any intellectual development. Without contradictions, without criticism, there would be no rational motive for changing theories: there would be no intellectual pro-gress. (Popper, 1963, p. 316) The major premise of the target article was that a metatheory, from a cognitive perspective, needs to emerge in the field of learning dis-

CSLI Language and Natural Reasoning The goal • To persuade you that one can do formal reasoning in natural language without translating it into a formal language, perhaps more than you thought possible. That’s the Natural Logic part. • To demonstrate that Natural Logic is not sufficient by itself

We congratulate the authors for a well written and thoughtful survey of some of the literature in this area. They are mainly concerned with the geometry and the computational learning aspects of the support vector machine (SVM). We will therefore complement their review by discussing from

for thoughtful comments. The paper has also beneÞted from comments of the

UNCTADs Division on Globalization and Development Strategy, for his thoughtful comments and