That, and if you had multiple series in a non-stacked Column chart, it would be difficult to (visually) tell the difference between data points which share the same date and those that are adjacent in time. You can simulate a "date" axis on your own, though, by inserting rows with null data in them in between your actual data points. You'll have to choose some level of granularity to represent the smallest measured unit of time in your data, and insert rows based on that. For example, if you charted data for the month of March, on a daily basis, your DataTable might look like this:
var data = new google.visualization.DataTable();
data.addColumn('string', 'Date');
data.addColumn('number', 'Foo');
data.addColumn('number', 'Bar');
data.addRows([
['3/1', 7, 3],
['3/2', null, null],
['3/3', null, null],
['3/4', 4, 8],
['3/5', 2, 9],
['3/6', 6, 7],
['3/7', null, null],
['3/8', 8, 2],
.
.
.
]);