By a ratio of nearly 2-to-1, Maine voters made clear Tuesday their objection to borrowing $25 million to complete the overhaul of Maine’s prison system. In the absence of any organized opposition, reasons for this rejection are unspecified, but they seem to be a combination of conflicting views: an aversion to spending money on criminals because they are criminals and an aversion to spending money on prisons because they are an inhumane way to treat human beings who have committed crimes. Add to those deeply held beliefs the fact that the primary beneficiary of the bond would have been Washington County, a place much of Maine finds easy to ignore, and the miracle of Question 1 is that one-third of voters supported it.

The defeat of Question 1 was not just resounding, it was preordained. History shows that stand-alone bond referendums for prisons never pass, that is why the successful ones always are bundled with other public-works projects likely to enjoy public support. So, initially, was this one: It began as part of a $43 million question that included homeland security and economic development projects. The decision made by lawmakers last spring to separate out the corrections component because it was not sufficiently related to the other two components was nonsensical given the lack of any relationship whatsoever between homeland security and economic development. It remains a mystery why lawmakers’ sudden interest in relatedness did not carry over to Question 2, the successful environmental bond that contained money for, in addition to wastewater treatment and tire stockpiles, the non-environmental matter of potato marketing.

What’s done, however, is done. What matters now is what’s done next. The need for this project did not disappear with the counting of ballots. Maine’s corrections system still needs the minimum-security pre-release center and the enhanced substance abuse programs (at Machias) and the expanded infirmary (at Windham) that this bond would have built. Maine taxpayers still need the increased cost-effectiveness of a modernized replacement for the dilapidated but amazingly efficient Downeast Correctional Facility. Washington County still needs the 70-plus jobs. And, like it or not, lawbreakers need the substance-abuse programs, the pre-release preparation and the medical services this bond would have provided.

One option, instead of building a new $13.9 million facility at Machias, is to spend several million rebuilding DCF’s substandard wastewater treatment system and continuing to live with the safety and efficiency issues that result from using a former military base as a prison. Considering that the new facility would have lowered the daily per-prisoner cost from $130 to $85, a 34 percent improvement, that is a poor solution.

Better that the incoming group of lawmakers, a new governor and a fresh batch of legislators, try again. All the rest of the massive corrections overhaul – more than $150 million to build a new state prison, plus facilities for women and juvenile offenders – was done without bond referendums. The same imaginative lease-purchase arrangements and borrowing through the Government Facilities Improvement Authority that made the previous improvements possible without referendums could be used again. If, however, lawmakers feel obligated to go to referendum, they must either bundle this with other, more popular projects or, if relatedness is the new standard, at least to be consistent about it.