This paper derives the characteristic function of the log-price in rough Heston models, where the volatility process is driven by a fractional Brownian motion with Hurst parameter H = α - 1/2 ∈ (0, 1/2). The authors use a novel approach based on nearly unstable Hawkes processes, which provide microstructural foundations for rough volatility. They construct a sequence of bi-dimensional Hawkes processes whose intensities converge to the rough Heston volatility dynamics. By analyzing the characteristic function of these point processes and taking a suitable limit, they obtain an explicit formula for the characteristic function of the log-price in the rough Heston model. The formula retains the exponential-affine structure of the classical Heston model, but the ordinary Riccati equation is replaced by a fractional Riccati equation involving a fractional derivative of order α. This fractional Riccati equation can be solved numerically using a predictor-corrector scheme, enabling efficient calibration and pricing. The results bridge the gap between rough volatility models and practical derivatives pricing, showing that the computational complexity remains manageable despite the non-Markovian nature of fractional volatility.