4 min read

How are ceasefire negotiations playing in Israeli and Arab media?

Sentiment around the ceasefire rose in both Israeli and Arabic-language media, but a more fine-grained analysis through Talisman showed just how far apart the two sides remain.
How are ceasefire negotiations playing in Israeli and Arab media?
Photo by Mohammed Ibrahim / Unsplash

A ceasefire proposal between Israel and Hamas hung in the balance this week. Hamas accepted the terms of the deal brokered by the Egyptians and Qataris, while Israel—which had offered a proposal of its own in late April—turned the Hamas version down. It  agreed to keep negotiating, but also declared its intention to move ahead with its planned Rafah operation. 

As news of the potential ceasefire broke, FilterLabs.AI used our Talisman data platform to track the story across Israeli and Arab-language media landscapes. Talisman allows us to dig into sentiment around the potential ceasefire—both in mainstream media sources and across a variety of social media platforms.

In the Israeli mainstream media, news of a potential ceasefire deal corresponded with a small but detectable rise in sentiment scores:

In the middle of April, many news stories reported that there had been little progress in the peace talks taking place in Egypt. When mainstream outlets started to report progress in the talks, around 4/25, sentiment in their coverage rose—but only modestly. The potential ceasefire seemed to be viewed positively, but failed to generate much enthusiasm. 

One of the most powerful capabilities of Talisman is the access it provides to the underlying artifacts, rather than merely the averaged sentiment score. When we used Talisman to dig into the discourse and get a closer look, we found that the Israeli media mainly portrayed Hamas as an ongoing obstacle to a lasting peace. Reports noted that 17 countries had called for Hamas to release Israeli hostages immediately, while the proposed peace agreement did not include a promise to release them.

On Israeli social media, the news of a potential ceasefire drew a mixed reaction. 

Sentiment on the subject had been far more volatile in Israeli online forums, messaging platforms, and social media than in the news (so different that we used two different scales to chart the trends). When Hamas announced its acceptance of the Egyptian-Qatari ceasefire proposal on May 6, sentiment in posts mentioning a ceasefire rose only modestly. And in examining some of the underlying artifacts, we found that people were not clamoring for the deal. Once again, the focus was on Hamas. According to posts on Telegram and other social platforms, the US was keeping pressure on Hamas, with Biden saying that Hamas was the only obstacle to a real peace deal. 

So although sentiment scores in stories about the ceasefire rose a bit after Hamas’s counter-proposal, a closer analysis of the artifacts through the Talisman platform revealed that Israeli commentators – whether in the mainstream media or online – were not too excited about the potential agreement.  

In the news and social media from Arabic-speaking countries, on the other hand, the reaction to the ceasefire proposal was extremely positive.

In Saudi Arabia, sentiment around a ceasefire rose more than 20 points in mainstream media sources:

It rose by nearly 50 points on social media:

A quick aside here about interpreting sentiment data: The astute reader may have detected that the Saudi sentiment scores are, on average, far lower than those in Israeli news and social media. That may be telling—but it may not. With sentiment data, the most important measure is changes in sentiment, not a given number itself. Because of differences in discourse style in different settings (such as news versus social discussion), and the unique scoring developed for each large language model (LLM), one must be cautious about drawing any conclusions from a discrepancy in scores between two different datasets, or even about the attitude on a topic from individual scores in one dataset. Still, a 50-plus point difference does at least suggest that discussion of a ceasefire in Saudi Arabia is accompanied by more negative emotion, on average, than conversations on the subject in Israel.

In both mainstream and social media discussions about a ceasefire, sentiment rose dramatically around April 26th. Using Talisman to get a more granular view of responses to the ceasefire proposal, FilterLabs found that Saudi sources were indeed enthusiastic. Mainstream Saudi media outlets included numerous statements from Hamas. Hamas senior leaders assured the Saudi readers that victory was in sight, that the ceasefire deal was a good one, and that they were grateful to Egypt and other Arab nations for their support. Sentiment rose much more in Saudi Arabia than it did in Israel. 

The same pattern we saw in Saudi Arabia held for most other Middle Eastern countries, including Egypt, the Emirates, Lebanon, and Palestine itself. The exact date that sentiment began to rise varied, likely influenced by other regional factors. For example, a number of the ceasefire-related posts Talisman unearthed in Egyptian and Palestinian media commemorated the anniversary of Egypt’s recovery of Sinai on April 25, and these contributed to the overall rise in sentiment. But throughout the region we found that sentiment around a ceasefire saw an uptick in late April and early May.  

Overall, FilterLabs found stark differences between Israeli and Arabic-language responses to the potential ceasefire. Sentiment around the ceasefire rose in both Israeli and Arabic-language media, but a more fine-grained analysis through Talisman showed just how far apart the two sides remain. Arab media was almost uniformly supportive of the proposal. Israelis, however, were far more ambivalent, both in their mainstream press and in individual discussions online. Given the lack of support from the press and population, it is no surprise that the Israeli government didn’t accept the May 6 ceasefire proposal. 

FilterLabs.AI is a data analytics company based in Cambridge, Massachusetts. We leverage natural language processing and tailored data modeling to scour and analyze global online communications and deliver hyper-local insights. Learn more at filterlabs.ai.