The latest fighting between Israel and the Hamas government in Gaza has now gone on for almost a month. In contrast to public opinion in much of the rest of the world, yet consistent with past outbreaks of Palestinian-Israeli violence, the American public consistently supports the Israeli government's policies. A Pew Research Center poll last week found that only 25% of Americans thought that the Israeli military response in Gaza had "gone too far," while 50% thought the response was "about right" or had "not gone far enough." In past Israeli military actions in 2006 and 2009, a very similar 23% and 24% of Americans, respectively, thought Israeli military actions had "gone too far." In the same Pew poll, 40% of Americans blamed Hamas for the current violence while only 19% blamed Israel.

​It is tempting to assume that the American public reacts to the facts on the ground, the specific policies pursued by the two sides, or the complicated series of events that led to the current conflict. But I think this is misguided. There are three things that are essential for understanding these poll results, none of which have anything to do with the details of the current tragic conflict.

First, most people don't follow the details of political news, including (perhaps especially) foreign conflicts that don't involve U.S. troops. And as with domestic political controversies, those most likely to be following this month's Israel-Hamas conflict closely are those with pre-existing strong views on the subject who will not change those views based on the specific details.

Second, in the absence of detailed knowledge of the topic, the large portion of the public that is inattentive to politics comes to an opinion based on group attachments. The strongest and most powerful group attachment is party identification. If the parties take clear positions on a topic, then their supporters in the mass public will follow those cues. However, as in past cases where violence has flared up between Israel and the Palestinians, national party politicians are not taking clearly divergent positions. There is wide support for Israeli policy in Gaza among Republican politicians, yet even among the most left-wing Democrats in Congress, it is almost impossible to find any willing to criticize the current Israeli military operation. As a result, even though Democrats are ideologically predisposed to be more skeptical of militarism generally, the partisan division of opinion on the Israeli military response is modest. Among Democrats, 35% think that the Israeli military response in Gaza has "gone too far," while 40% think it is "about right" or has "not gone far enough." ​

What shapes opinions among the inattentive public in the absence of clear party messages? This brings me to my third point. The most powerful influence on mass opinion after partisanship is feelings toward different ethnic and religious groups. The connection between ethnic and racial attitudes and political opinions is well documented. And Adam Berinsky has shown that ethnic attachments can have important effects even on attitudes toward foreign conflicts. Here is where opinions on the Gaza conflict become more understandable. A June Pew poll asked Americans to rate various religions on a 1-100 "feeling thermometer" scale. As you can see, in the U.S., Jews are the most popular religious group, with an average thermometer rating of 62, while Muslims are the least popular, with an average thermometer rating of 40. To be sure, the ratings of Jews are not statistically distinguishable from Catholics and evangelical Christians, and the ratings of Muslims are not distinguishable from those of atheists. But the important thing is the huge gap between feelings toward Jews and Muslims.

A Zogby poll sponsored by the Arab American Institute found very similar results. It polled favorability ratings for various religious and ethnic groups and found that Jews were the most popular group, with a 66% overall favorability rating, while Arabs and Muslims were the two least popular groups, with 32% and 27% favorability ratings, respectively. The Zogby poll breaks this down by partisanship and finds very little difference in how Democrats, Republicans and Independents rate Jews, Arabs and Muslims.

Consistent with this connection between conflict attitudes and group attitudes, young Americans both hold the least negative attitudes toward Arabs and Muslims and are the least likely to blame Hamas alone for the current violence.

To a large portion of the American public, the details of any military confrontation between Israel and the Palestinians are too obscure to affect their thinking. Their first instinct would be to rely on partisan cues. Among Republicans, elite partisan cues clearly favor Israeli policy. But as Democratic politicians are either supportive of Israeli policy, dodge the issue or don't take a clear stand, substantial numbers of Democrats as well as Independents may not have clear partisan cues to follow. In the absence of partisanship, ethnic and religious groups are the second major determinant of political preferences among the less politically engaged public. For these people who don't know all the details, the most basic facts are that this is a conflict between a Jewish state and the Arab and Muslim Palestinians. (A small but nontrivial portion of Palestinians are actually Christian, but this is not widely known to Americans.) These attitudes toward groups will push people strongly toward supporting Israel's position in whatever conflict is taking place.

​I think there are two important take-aways from all of this. First, because Americans are so inattentive to the details of political controversies, and hold such consistent views on every Israeli-Palestinian violent clash, we shouldn't see their opinions as a reflection on the details of any specific clash. The American public's endorsement of current Israeli policy largely isn't a reaction to that policy because most people aren't following the details at all.

Second, the one thing that could change U.S. mass opinion would be if party elites changed their messages. The one group attachment powerful enough to potentially overwhelm group attitudes is party identification. For instance, if most Democratic politicians in Washington came out against an Israeli military operation, that could potentially lead ordinary Democrats to follow those cues rather than their group attitudes when forming an opinion. If that happened, American mass opinion would likely become much more split than it is today. Just as opinions on the conflict today are not determined by the public's assessments of the facts on the ground, opinion would potentially change in the future not because the mass public came to a different assessment of the facts, but because political elites decided to change their positions.