Just to be clear, I was suggesting that in a very small number of difficult cases, level crossings might be justifiable to avoid enormous expense, likely great local opposition to a bridge, or a long rail deviation perhaps taking the new railway and its local station a long way out of a settlement it could otherwise serve more optimally. The case of a traditional station crossing near housing in a town/village setting comes to mind, where trains would also be speed limited by a station approach or start from rest, allowing a locally monitored solution to be considered without major effect on journey times. For crossings of all sorts on rural plain line sections where trains need to be going full speed, I would agree that bridging should always be the default option for multiple reasons. I fully agree crossings are a pain to manage. There's always some new modification to do on circuits and signage and periodic tests and inspections, etc, etc. And they're an open (literally) invitations to trespassers. By contrast a bridge is almost 'fit and forget', but there are some locations where bridges very close to the old crossing site would be impossible and a complete road closure is undesirable and likely unachievable because the road provides some sole or most convenient access. There are many cases like this on the existing railway where long term full crossing abolition is not reasonably practicable for various reasons. I don't see why new railways shouldn't also have SOME new crossings in similar environments, properly risk mitigated of course and notwithstanding a continued commitment to progressively reduce the total number of crossings networkwide. In short, I'm trying to suggest, generically, the 'exceptional circumstance' that some new crossings might be allowed to be provided in.