ABSTRACT
“Where is the Palestinian Gandhi?” is a question often asked in relation to the nature of Palestinian resistance. The question reflects a lack of accurate knowledge about Palestinian society and history, as well as a misrepresentation of the forms in which Palestinian resistance has taken place over the past few decades. Throughout their struggle for independence, Palestinians have predominantly adopted nonviolent forms of resistance. Focusing on the determinants of Palestinian support for nonviolent resistance against Israeli occupation two decades after the signing of the Oslo Accords in 1993, this paper lays out the historical context of nonviolence in the Palestinian political landscape. Through the use of survey data this study provides an alternative theoretical conceptualization of Palestinian popular resistance, and community resilience. The findings of this empirical study highlight the impact of self-reported identities on support for nonviolent tactics in resisting Israeli occupation.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes
1 Survey conducted in the summer of 2013, to coincide with the 20th anniversary of the signing of the Oslo accords. N = 832. Study conducted in the West Bank only. Attempts to carry out the survey in the Gaza Strip were unsuccessful, due to the security situation.
2 For example, stone throwing at tanks/military vehicles is widely considered to be a nonviolent act due to the relatively low harm the act would cause, but in September 2015, Israeli Knesset (Parliament) passed a law further criminalizing stone throwing at Israeli soldiers, by setting the new punishment at 20 years in prison. For the purposes of this article, I focus on the asymmetrical power structure to guide this definition.
3 For more information on the Boycott, Divestment, and Sanctions (BDS) campaign, its origins and nature, refer to Omar Barghouti’s Boycott, Divestment, Sanctions: The Global Struggle for Palestinian Rights. 2011. Haymarket Books.
4 Ordered logistic regression assumes that the relationship between all covariates in the model and each category of the dependent variable are equal. This is the parallel regression assumption. The Brant test is commonly used to assess whether or not this assumption is violated within the model. In this case the Brant test could not be completed due to low cell counts in the categories 0 and 1 in each of the dependent variables of interest. In the event that the parallel regression assumption is violated, multinomial regression is the recommended alternative model. I therefore conducted additional analysis using multinomial logistic regression. The findings generated by the multinomial model largely corroborate those generated by the ordered logistic regression across all three dependent variables, where those identifying as Palestinian and Arab were more likely to highly rate non-violence as an effective means of dealing with the occupation than their Islamic counterparts, and those who identified as Arab were more likely than those who identify as Islamic to highly rate lobbying and BDS. As an added step, I modeled the data using a Poisson regression, treating the outcome like count data. The dependent variables each have a mean of about three with a standard deviation of about 1.3. Again the findings corroborate those derived from the ordered logistic regression model and the multinomial logistic regression. Since this is the case, and neither a multinomial or Poisson regression model are theoretically a good fit for the data, I continue the analysis using an ordered logistic regression model.