Recent experiments show the strong water uptake of gas shales which are strongly oil-wet based on contact angle measurements.(1,2) Clay hydration, microfracture induction, lamination, and osmotic effect are collectively responsible for the excess water uptake. However, the previous measurements are not sufficient to isolate the above factors nor to explain why the bulk of shale samples can hardly imbibe the oil which completely spreads on their surface. To answer the remaining questions, we measure and compare spontaneous imbibition of oil and water into the crushed packs of the similar shales. In contrast to the intact samples, the crushed samples consistently imbibe more oil than water. The comparative study suggests that the connected pore network of the intact samples is water wet while the majority of rock including poorly connected pores is oil-wet. This argument is backed by complete spreading of oil on fresh surfaces of the rock. In contrast to the artificial pores of crushed rock, the existing pores of intact rock are already wetted by a film of water and/or covered by precipitated salt, which gives the pores a preference for water over oil. Furthermore, the presence of salt in the pore space provides an additional force for water uptake through an osmotic effect. This argument is backed by the observed reduction in shale alteration and water imbibition through increasing the salt concentration.