I've come to beleive the operational level of war is the linking of a series of objectives as they relate to a desired (strategic) end, and the arranging of resources and capabilities in time and space to achieve those objectives. The tactical actions then are those understood to be required to acheive said operational objectives, or those that occur as a result of pursuing the operational objective.

Those tactical sctions may not always be the ones you anicpated, but if there is resistance then it stands to reason that some type of tactical action will be required. I'd also note that loss of a given tactical action does not mean the end cannot be fulfilled, or even that some part of the operational objective may not be fulfilled, but tactical success does make a positive differnce (as long as it does not operationally exhaust you).

For a great example I think you could look at Grant's Wilderness campaign, and consider "Cold Harbor". Certainly a bad tactical outcome for Union Forces, however Grant was able to continue afterwards and soon secured City Point, set up a masive LOG base, and then contained Lee's Forces (as well as forcing the Confederate govt to flee Richmond. Grant also fulfilled the strategic end of destroying the Army of N. Virginia, occupying the Conf. capital and getting Lee to surrender.

If all of Grant's tactical actions had ended in failure that halted his operational momentum, or allowed Lee's army to escape, a subsequent campaign may have been required.

ref. the article Wilf brought up, I think its great for discussion. I think there can be a real danger in it happening, but historically it seems most likely when strategic concepts and defined policy end are absent or ambiguous (it could be we just don't really know what we want, or prefer to keep options open) - however, we tend to be action oriented, and there also seems to be some basis for expectations for us to be so.

Best, Rob