The recent deaths and illnesses from pasteurized-milk ice cream emphasize, yet again, how much less we know about food safety than the people in charge would care to admit.

In the span of a few months, five ice cream companies have recalled potentially tainted products. Late last year, it was Snoqualmie Ice Cream and Pink’s Ice Cream out of Washington state, followed by Full Tilt Ice Cream, also of Washington— all based on the possibility of listeria monocytogenes in their ice cream and production facilities.

Then it was Blue Bell Creameries of Texas, the third-largest producer nationally, in which three people died from listeria monocytogenes in its ice cream (along with at least seven other illnesses, according to the U.S. Centers for Disease Control).

Most recently, Jeni’s Ice Cream of Ohio decided to do a recall and shut down shops after a U.S. Department of Agriculture test found listeria monocytogenes in its ice cream.

The professionals who monitor food safety have contradicted their own warnings about ice cream. The Center for Science and the Public Interest, a nonprofit organization that tends to parrot the U.S. Centers for Disease Control, in a 2009 report included ice cream as one of America’s ten most dangerous foods, citing nearly 75 outbreaks in the 20 years from 1990 to 2009. “These scoops can occasionally carry a load of dangerous bacteria,” it said. (It also included another dairy product in its top-ten list–cheese.)

But then, in a 2014 report, it labeled dairy as one of our safest food categories. (“Pound-for-pound, seafood is the riskiest food, followed by poultry; fruit and dairy are the safest foods,” CSPI concluded.)

The same kinds of contradictions crop up when the professionals consider food safety solutions. Here are a few of the solutions tossed around:

More mandated testing.

One seemingly obvious lesson seems that we should do more testing of dairy products for listeria monocytogenes than the occasional random testing that now occurs. Unfortunately, the presence of pathogens doesn’t necessarily equate with the development of illnesses.

Part of the problem is that trying to find the pathogens in food that will cause disease isn’t unlike trying to find the needle in the proverbial haystack. The New York Department of Agriculture and Markets has shut down raw milk dairies on at least a couple dozen occasions over the last decade for the presence of listeria monocytogenes, without a single illness developing (either in New York, or anywhere else in the country).

The U.S. Food and Drug Administration spent much of 2014 doing what it called a “pilot testing program” of more than 1,600 samples of raw milk cheese from dozens of producers in the U.S. as well as producers from nearly 20 other countries. It found listeria in the raw-milk cheese of only one American producer, so far as is known, and forced that producer to recall all the cheese; no illnesses were reported, even though much of the cheese had almost certainly been consumed before the recall. (The FDA also found listeria and salmonella in five imported cheese samples, and no illnesses were reported in those cases as well.)

Tighten the regulations.

The U.S. actually has the tightest regulations in the world when it comes to listeria monocytogenes in foods—a zero tolerance policy. Thus, the presence of a single cell of listeria is enough to shutter a food producer, even if no illnesses have occurred.

Because the presence of just a few cells of listeria monocytogenes doesn’t usually lead to illness, most First World countries, including Canada and members of the European Union have adopted policies whereby they accept the presence of small amounts of listeria without shutting food producers.

Do quicker recalls when pathogens show up.

No matter how quickly a company orders a product recall, the reality is that most food is consumed shortly after it is purchased. All the ice cream recalls this year very likely came after much of the questionable ice cream had been eaten.

More aggressive anti-pathogen treatments of milk.

Some in the food safety community have promoted the idea of more intense pasteurization, heating the milk even further than the 165 degrees (F) regular pasteurization takes place at.

Yet researchers from Cornell University last year threw cold water on that idea in a paper considering the public health effects of raising the pasteurization temperature about 10 per cent, as a way to counter possible bioterrorism. Its conclusion? “…annual listeriosis deaths from consumption of this milk would increase from 18 to 670, a 38-fold increase…”

Punish companies that don’t rid their facilities of listeria.

In practice, once listeria infects a food production facility, it is extremely difficult to eliminate. Blue Bell Creameries has been unable to re-open for weeks now because listeria keeps showing up in tests of its plant facilities.

It’s safe to say the owners of Blue Bell want more than anything to rid their facilities of listeria monocytogenes, But listeria is extremely persistent, and eliminating every last cell is quite difficult, and sometimes apparently impossible, short of dynamiting the facility.

The solution isn’t to ban ice cream, any more than it is to ban petting zoos or raw milk. The solution is to be seeking out ways to reduce risk, both in food production and consumption habits. It could be the petting zoo concept needs re-thinking, so animals are less stressed, and able to live more naturally. Ice cream producers likely need more education about ways to keep their plants in safe order.

But though risk can be reduced, its uncertainties will never be eliminated. And that I can say for certain.